Package 'HH'

Description

Support software for Statistical Analysis and Data Display (Second Edition, Springer, ISBN 978-1-4939-2121-8, 2015) and (First Edition, Springer, ISBN 0-387-40270-5, 2004) by Richard M. Heiberger and Burt Holland. This contemporary presentation of statistical methods features extensive use of graphical displays for exploring data and for displaying the analysis. The second edition includes redesigned graphics and additional chapters. The authors emphasize how to construct and interpret graphs, discuss principles of graphical design, and show how accompanying traditional tabular results are used to confirm the visual impressions derived directly from the graphs. Many of the graphical formats are novel and appear here for the first time in print. All chapters have exercises. All functions introduced in the book are in the package. R code for all examples, both graphs and tables, in the book is included in the scripts directory of the package.

Details

The DESCRIPTION file:

Index of help topics:

AEdotplot               AE (Adverse Events) dotplot of incidence and
                        relative risk
AEdotplot.data.frame    AE (Adverse Events) dotplot of incidence and
                        relative risk, support functions
CIplot                  Illustration of the meaning of confidence
                        levels.
Discrete4               Discrete with four levels color dataset.
EmphasizeVerticalPanels
                        Helper function for likertWeighted(). used for
                        vertical spacing and horizontal borders of
                        grouped panels.
F.curve                 plot a chisquare or a F-curve.
GSremove                Remove selected GraphSheetPages in the S-Plus
                        Windows GUI Graphsheet
HH-package              Statistical Analysis and Data Display:
                        Heiberger and Holland
HH.regsubsets           Display tabular results for Best Subsets
                        Regression.
HHscriptnames           Find absolute pathname of a script file for the
                        HH book in the HH package.
InsertVerticalPanels    Expand a 3D array on the second dimension,
                        inserting empty layers where the input vector
                        has a '0' value.  A 2D argument 'x' with
                        'dim(x)==c(r,c)' is first extended to 3D with
                        'dim(x)==c(1,r,c)', and then the result is
                        collapsed back to 2D.
LikertPercentCountColumns
                        Display likert plots with percents in the first
                        column of panels and counts in the second
                        column of panels.
NTplot                  Specify plots to illustrate Normal and t
                        Hypothesis Tests or Confidence Intervals,
                        including normal approximation to the binomial.
NormalAndTPower         Construct a power graph based on the NTplot.
NormalAndTplot          Specify plots to illustrate Normal and t
                        Hypothesis Tests or Confidence Intervals.
OddsRatio               Calculate or plot the odds ratio for a 2x2
                        table of counts.
OneWayVarPlot           Displays a three-panel 'bwplot' of the data by
                        group, of the group means, and of the entire
                        dataset.  This is an approximate visualization
                        of the Mean Square lines from the ANOVA table
                        for a one-way ANOVA model.
ResizeEtc               Display multiple independent trellis objects on
                        the same coordinated scale.
ResizeEtc.likertPlot    Display multiple independent trellis objects,
                        representing likert plots, on the same
                        coordinated scale.
ToBW.likert             Change colors in a likert plot to shades of
                        Black and White.
X.residuals             Residuals from the regression of each column of
                        a data.frame against all the other columns.
ae.dotplot              AE (Adverse Events) dotplot of incidence and
                        relative risk
ancova                  Compute and plot oneway analysis of covariance
ancova-class            Class "ancova" Analysis of Covariance
ancovaplot              Analysis of Covariance Plots
anova.ancovaplot        ANOVA table for a c("ancovaplot","trellis")
                        object.
anovaMean               ANOVA table from the group sample sizes, means,
                        and standard deviations.
aovSufficient           Analysis of variance from sufficient statistics
                        for groups.
arima.diag.hh           Repair design error in S-Plus arima.diag
arma.loop               Loop through a series of ARIMA models and
                        display coordinated tables and diagnoastic
                        graphs.
as.likert               Support functions for diverging stacked
                        barcharts for Likert, semantic differential,
                        and rating scale data.
as.matrix.listOfNamedMatrices
                        Convert a list of numeric matrices to a single
                        matrix
as.multicomp            Support functions in R for MMC (mean-mean
                        multiple comparisons) plots.
as.rts                  Miscellaneous functions that I wish were in or
                        consistent between S-Plus and R.
as.vector.trellis       Convert a two-dimensional trellis object into a
                        one-dimensional trellis object. Change the
                        order of panels in a trellis object.
axis.i2wt               specialized axis function for interaction2wt.
bivariateNormal         Plot the bivariate normal density using
                        wireframe for specified rho.
case                    case statistics for regression analysis
ci.plot                 Plot confidence and prediction intervals for
                        simple linear regression
col.hh                  Initializing Trellis Displays
col3x2                  col3x2 color dataset
combineLimits.trellisvector
                        Combine limits on a one-dimensional trellis
                        object.
cp.calc                 Rearranges and improves the legibility of the
                        output from the stepwise function in S-Plus.
cplx                    Generate a sequence spanning the xlim of a
                        lattice window.
datasets                Datasets for Statistical Analysis and Data
                        Display, Heiberger and Holland
dchisq.intermediate     Intermediate f and chisq functions to simplify
                        writing for both R and S-Plus.
diag.maybe.null         Returns a value for the diagonal of NA and NULL
                        arguments.
diagQQ                  QQ plot of regression residuals.
diagplot5new            Transpose of ECDF for centered fitted values
                        and residuals from a linear model.
do.formula.trellis.xysplom
                        Interprets model formulas for xysplom and
                        extended bwplots
emptyMainLeftAxisLeftStripBottomLegend
                        Remove main title, left axis tick labels, left
                        strip, bottom legend from plot and keep the
                        vertical spacing allocated to those items.
export.eps              Exports a graph to an EPS file.
glhtWithMCP.993         Retain averaging behavior that was previously
                        available in glht.
gof.calculation         Calculate Box-Ljung Goodness of Fit for ARIMA
                        models in S-Plus.
grid.yaxis.hh           make x- and y-axis labels
hhpdf                   R tools for writing HH2: hhpdf, hhdev.off,
                        hhcapture, hhcode, hhpng, hhlatex
hov                     Homogeneity of Variance
hovBF                   Homogeneity of Variance: Brown-Forsyth method
hovPlot                 Homogeneity of Variance Plot
if.R                    Conditional Execution for R or S-Plus
interaction.positioned
                        interaction method for positioned factors.
interaction2wt          Plot all main effects and twoway interactions
                        in a multifactor design
interval                Prediction and Confidence Intervals for glm
                        Objects
intxplot                Interaction plot, with an option to print
                        standard error bars.
ladder                  Draw a "ladder of powers" plot, plotting each
                        of several powers of y against the same powers
                        of x.
latex.array             Generate the latex code for an '"array"' or
                        '"table"' with 3, 4, or more dimensions.
latticeresids           Subroutine used by residual.plots.lattice
legendGrob2wt           place separate keys to the left of each row of
                        a trellis
likert                  Diverging stacked barcharts for Likert,
                        semantic differential, rating scale data, and
                        population pyramids.
likertColor             Selection of colors for Likert plots.
likertMosaic            Diverging stacked barcharts for Likert,
                        semantic differential, rating scale data, and
                        population pyramids based on mosaic as the
                        plotting style.
likertWeighted          Special case wrapper for likert() when multiple
                        columns are to have the same bar thicknesses.
                        Uses formula with one or two conditioning
                        variables.
lm.regsubsets           Evaluate lm model with highest adjusted $R^2$.
lmatPairwise            lmatPairwise
lmatRows                Find the row numbers in the lmat corresponding
                        to the focus factor.
lmplot                  Four types of residual plots for linear models.
logit                   Logistic and odds functions and their inverses.
matrix.trellis          Convert a one-dimensional trellis object to a
                        two-dimensional trellis object. This permits
                        combineLimits and useOuterStrips to work.
mcalinfct               MCA multiple comparisons analysis (pairwise)
mmc                     MMC (Mean-mean Multiple Comparisons) plots.
mmc.mean                MMC (Mean-mean Multiple Comparisons) plots from
                        the sufficient statistics for a one-way design.
mmcAspect               Control aspect ratio in MMC plots to maintain
                        isomeans grid as a square.
mmcPruneIsomeans        MMC plots in lattice-suppress isomeans grid
                        lines for specified levels of the factor.
mmcisomeans             Functions used by mmcplot.
mmcplot                 MMC (Mean-mean Multiple Comparisons) plots in
                        lattice.
multicomp.order         Update a multicomp object by ordering its
                        contrasts.
multicomp.reverse       Force all comparisons in a "multicomp" object
                        to have the same sign.
norm.curve              plot a normal or a t-curve with both x and z
                        axes.
normalApproxBinomial    Plots to illustrate Normal Approximation to the
                        Binomial-hypothesis tests or confidence
                        intervals.
npar.arma               Count the number of parameters in an ARIMA
                        model specification.
objip                   loop through all attached directories looking
                        for pattern, possibly restricting to specified
                        class or mode.
orthog.complete         Construct an orthogonal matrix which is an
                        arbitrary completion of the column space of the
                        input set of columns.
panel.acf               Panel functions for tsdiagplot.
panel.axis.right        Right-justify right-axis tick labels.
panel.bwplot.intermediate.hh
                        Panel functions for bwplot.
panel.bwplot.superpose
                        Panel function for bwplot that displays an
                        entire box in the colors coded by groups.
panel.bwplott           Extension to S-Plus trellis to allow transposed
                        plots.
panel.cartesian         trellis panel function, with labeled rows and
                        columns and without strip labels.
panel.ci.plot           Default Panel Function for ci.plot
panel.confintMMC        Confidence interval panel for MMC tiebreaker
                        plots, or confidence interval plot.
panel.dotplot.tb        Dotplot with evenly spaced tiebreakers.
panel.interaction2wt    Plot all main effects and twoway interactions
                        in a multifactor design
panel.isomeans          isomeans grid for MMC plots.
panel.likert            Panel functions for likert that include a
                        stackWidth argument
panel.pairs.hh          Function based on S-Plus panel.pairs to add the
                        subpanel.scales and panel.cex arguments.
panel.xysplom           panel method for xysplom.
partial.corr            partial correlations
pdf.latex               Construct a pdf file from a "latex" file.  See
                        Hmisc::latex for concepts.
pdiscunif               Discrete Uniform Distribution
perspPlane              Helper functions for regr2.plot
plot.mmc.multicomp      MMC (Mean-mean Multiple Comparisons) plot.
plot.multicomp          Multiple comparisons plot that gives
                        independent user control over the appearance of
                        the significant and not significant
                        comparisons.
position                Find or assign the implied position for
                        graphing the levels of a factor.  A new class
                        "positioned", which inherits from "ordered" and
                        "factor", is defined.
positioned-class        Class "positioned", extends "ordered" to
                        specify the position for graphing the levels of
                        a factor.
print.NormalAndTplot    Print method for Normal and t plots from
                        NTplot.
print.TwoTrellisColumns5
                        Print two conformable trellis plots in adjacent
                        columns with user control of widths.
print.latticeresids     Print a 'latticeresids' object.
print.tsdiagplot        Print a "tsdiagplot" object.
push.vp.hh              push and pop a grid viewport, turn clipping
                        off, change scale.
pyramidLikert           Print a Likert plot as a Population Triangle
rbind.trellis           Extend matrix reshaping functions to trellis
                        objects.
regr1.plot              plot x and y, with optional straight line fit
                        and display of squared residuals
regr2.plot              3D plot of z against x and y, with regression
                        plane fit and display of squared residuals.
regrresidplot           Draw a plot of y vs x from a linear model
                        object, with residuals indicated by lines or
                        squares.
resid.squares           plot squared residuals in inches to match the
                        y-dimension
residVSfitted           Draw plots of resid ~ y.hat and
                        sqrt(abs(resid)) ~ y.hat
residual.plots          Residual plots for a linear model.
residual.plots.lattice
                        Construct four sets of regression plots: Y
                        against X, residuals against X, partial
                        residuals against X, partial residuals against
                        each X adjusted for all the other X columns.
rowPcts                 Row and columns percents
seqplot                 Time series plot.
seqplotForecast         seqplot with confidence bands for the forecast
                        region.
strip.background0       Turn off the coloring in the trellis strip
                        labels.  Color 0 is the background color.
strip.useOuterStrips.first
                        Functions based on strip.default for use with
                        the useOuterScales function.
strip.xysplom           strip function that is able to place the
                        correlation or regression coefficient into the
                        strip label.
sufficient              Calculates the mean, standard deviation, and
                        number of observations in each group of a
                        data.frame that has one continuous variable and
                        two factors.
summary.arma.loop       summary and print and subscript methods for
                        tsdiagplot and related objects.
toCQxR                  Reshape a 3-way array to a 2-way data.frame
                        that can can be used with a trellis
                        conditioning formula to get the three-way
                        behavior. Used with likertWeighted().
tsacfplots              Coordinated time series and ACF and PCF plots.
tsdiagplot              Times series diagnostic plots for a structured
                        set of ARIMA models.
useOuterScales          Put scales for axes only on the bottom and left
                        panels of a lattice display, and give fine
                        control over the placement of strips
useOuterStripsT2L1      Three-factor generalization of
                        latticeExtra::useOuterStrips
vif                     Calculate the Variance Inflation Factor
xysplom                 scatterplot matrix with potentially different
                        sets of variables on the rows and columns.
z.test                  Z test for known population standard deviation

data display, scatterplot matrix, (MMC Mean–mean Multiple Comparison) plots, interaction plots, ANCOVA plots, regression diagnostics, time series, ARIMA models, boxplots

Author(s)

Richard M. Heiberger

Maintainer: Richard M. Heiberger <[email protected]>

References

Heiberger, Richard M. and Holland, Burt (2015). Statistical Analysis and Data Display: An Intermediate Course with Examples in R. Second Edition. Springer-Verlag, New York. https://link.springer.com/book/10.1007/978-1-4939-2122-5

Heiberger, Richard M. and Holland, Burt (2004). Statistical Analysis and Data Display: An Intermediate Course with Examples in S-Plus, R, and SAS, First Edition. Springer Texts in Statistics. Springer. https://link.springer.com/book/10.1007/978-1-4757-4284-8.

See Also

ancovaplot, ci.plot, interaction2wt, ladder, case.lm, NTplot for Normal and $t$ plots, hov, resid.squares, MMC, AEdotplot, likert, tsacfplots, tsdiagplot

demo(package="HH")

Examples

## In addition to the examples for each function,
## there are seven interactive shiny apps in the HH package:
## Not run: 
if (interactive())   NTplot(mean0=0, mean1=1, shiny=TRUE)
if (interactive())   shiny::runApp(system.file("shiny/bivariateNormal", package="HH"))
if (interactive())   shiny::runApp(system.file("shiny/bivariateNormalScatterplot", package="HH"))
if (interactive())   shiny::runApp(system.file("shiny/PopulationPyramid", package="HH"))
if (interactive())   shiny.CIplot(height = "auto")
if (interactive())   shiny::runApp(system.file("shiny/AEdotplot", package="HH"))
if (interactive())   shiny::runApp(system.file("shiny/likert", package="HH"))

## End(Not run)

Description

A two-panel display of the most frequently occurring AEs in the active arm of a clinical study. The first panel displays their incidence by treatment group, with different symbols for each group. The second panel displays the relative risk of an event on the active arm relative to the placebo arm, with 95% confidence intervals for a $2\times2$ table. By default, the AEs are ordered by relative risk so that events with the largest increases in risk for the active treatment are prominent at the top of the display. See the Details section for information on changing the sort order.

Usage

ae.dotplot(ae, ...)

ae.dotplot.long(xr,
                A.name = levels(xr$RAND)[1], B.name = levels(xr$RAND)[2],
                col.AB = c("red","blue"), pch.AB = c(16, 17),
                main.title = paste("Most Frequent On-Therapy Adverse Events",
                                   "Sorted by Relative Risk"),
                main.cex = 1,
                cex.AB.points = NULL, cex.AB.y.scale = 0.6,
                position.left = c(0, 0, 0.7, 1), position.right = c(0.61, 0, 0.98, 1),
                key.y = -0.2, CI.percent=95)

logrelrisk(ae, A.name, B.name, crit.value=1.96)

panel.ae.leftplot(x, y, groups, col.AB, ...)

panel.ae.rightplot(x, y, ..., lwd=6, lower, upper, cex=.7)

panel.ae.dotplot(x, y, groups, ..., col.AB, pch.AB, lower, upper) ## R only

aeReshapeToLong(aewide)

Arguments

Details

The second panel shows relative risk of an event on the active arm (treatment B) relative to the placebo arm (treatment A), with 95% confidence intervals for a $2\times2$ table. Confidence intervals on the log relative risk are calculated using the asymptotic standard error formula given as Equation 3.18 in Agresti A., Categorical Data Analysis. Wiley: New York, 1990.

By default the ae.dotplot function sorts the events by relative risk. To change the sort order, you must redefine the ordering of the ordered factor PREF. See the examples below.

Value

logrelrisk takes an input data.frame of the form x described in the argument list and returns a data.frame consisting of the input argument with additional columns as described in the argument xr. The result column of symptom names PREF is an ordered factor, with the order specified by the relative risk.

ae.leftplot returns a "trellis" object containing a horizontal dotplot of the percents against each of the symptom names.

ae.rightplot returns a "trellis" object containing a horizontal plot on the log scale of the relative risk confidence intervals against each of the symptom names.

ae.dotplot calls both ae.leftplot and ae.rightplot and combines their plots into a single display with a single set of left axis labels, a main title, and a key. The value returned invisibly is a list of the full left trellis object and the right trellis object with its left labels blanked out. Printing the value will not usually be interesting as the main title and key are not included. It is better to call ae.dotplot directly, perhaps with a change in some of the positioning arguments.

Author(s)

Richard M. Heiberger <[email protected]>

References

Ohad Amit, Richard M. Heiberger, and Peter W. Lane. (2008) “Graphical Approaches to the Analysis of Safety Data from Clinical Trials”. Pharmaceutical Statistics, 7, 1, 20–35. https://onlinelibrary.wiley.com/doi/10.1002/pst.254

See Also

AEdotplot for a three-panel version that also has an associated shiny app.

Examples

## variable names in the input data.frame aeanonym
## RAND   treatment as randomized
## PREF   adverse event symptom name
## SN     number of patients in treatment group
## SAE    number of patients  in each group for whom the event PREF was observed
##
## Input sort order is PREF/RAND

data(aeanonym)
head(aeanonym)

## Calculate log relative risk and confidence intervals (95% by default).
## logrelrisk sets the sort order for PREF to match the relative risk.
aeanonymr <- logrelrisk(aeanonym) ## sorts by relative risk
head(aeanonymr)

## construct and print plot on current graphics device
ae.dotplot(aeanonymr,
           A.name="TREATMENT A (N=216)",
           B.name="TREATMENT B (N=431)")
## export.eps(h2("stdt/figure/aerelrisk.eps"))
## This looks great on screen and exports badly to eps.
## We recommend drawing this plot directly to the postscript device:
##
## trellis.device(postscript, color=TRUE, horizontal=TRUE,
##                colors=ps.colors.rgb[
##                  c("black", "blue", "red", "green",
##                    "yellow", "cyan","magenta","brown"),],
##                onefile=FALSE, print.it=FALSE,
##                file=h2("stdt/figure/aerelrisk.ps"))
## ae.dotplot(aeanonymr,
##            A.name="TREATMENT A (N=216)",
##            B.name="TREATMENT B (N=431)")
## dev.off()

## To change the sort order, redefine the PREF factor.
## For this example, to plot alphabetically, use the statement
aeanonymr$PREF <- ordered(aeanonymr$PREF, levels=sort(levels(aeanonymr$PREF)))
ae.dotplot(aeanonymr,
           A.name="TREATMENT A (N=216)",
           B.name="TREATMENT B (N=431)",
           main.title="change the main title to reflect the new sort order")

## Not run: 
## to restore the order back to the default, use
relrisk <- aeanonymr[seq(1, nrow(aeanonymr), 2), "relrisk"]
PREF <- unique(aeanonymr$PREF)
aeanonymr$PREF <- ordered(aeanonymr$PREF, levels=PREF[order(relrisk)])
ae.dotplot(aeanonymr,
           A.name="TREATMENT A (N=216)",
           B.name="TREATMENT B (N=431)",
           main.title="back to the original sort order")

## smaller artifical example with the wide format
aewide <- data.frame(Event=letters[1:6],
                     N.A=c(50,50,50,50,50,50),
                     N.B=c(90,90,90,90,90,90),
                     AE.A=2*(1:6),
                     AE.B=1:6)
aewtol <- aeReshapeToLong(aewide)
xr <- logrelrisk(aewtol)
ae.dotplot(xr)

## End(Not run)

Description

A three-panel display of the most frequently occurring AEs in the active arm of a clinical study. The first panel displays their incidence by treatment group, with different symbols for each group. The second panel displays the relative risk of an event on the active arm relative to the placebo arm, with 95% confidence intervals for a $2\times2$ table. By default, the AEs are ordered by relative risk so that events with the largest increases in risk for the active treatment are prominent at the top of the display. By setting the argument sortbyRelativeRisk=FALSE, the AEs retain the order specified by the levels of the factor. The third panel displays the numerical values of number of patients for each treatment, number of adverse events for each treatment, and relative risk. The third panel can be suppressed by the print method.

Usage

AEdotplot(xr, ...)

  ## S3 method for class 'formula'
AEdotplot(xr, groups=NULL, data=NULL,
          sortbyRelativeRisk=TRUE,
          ...,
          sub=list(deparse(this.call[1:4],
                                width.cutoff=500), cex=.7))

Arguments

Details

The first panel is an ordinary dotplot of the percent of AE observed for each treatment by AE.

The second panel shows relative risk of an event on the Treatment B arm (usually the active compound) relative to the Treatment A arm (usually the placebo), with 95% confidence intervals for a $2\times2$ table. Confidence intervals on the log relative risk are calculated using the asymptotic standard error formula given as Equation 3.18 in Agresti A., Categorical Data Analysis. Wiley: New York, 1990.

By default the AEdotplot function sorts the events by relative risk. To retain the sort order implied by the levels of the AE factor, specify the argument sortbyRelativeRisk=FALSE. To control the sort order, make the AE factor in the input dataset an ordered factor and specify the levels in the order you want.

The third panel shows the numerical values of the number and percent of observed events on each arm and the relative risk. The display of third panel can be suppressed by specifying the panel.widths argument. See the discussion of the panel.widths in AEdotplot.data.frame.

Value

The primary interest is in the display of the plot.

The function returns an AEdotplot object which is a list of three trellis objects, one for the the Percent plot, one for the Relative Risk plot, and one for the Text plot containing the table of input values. The object has attributes

1. main and sub hold the main and subtitles. Each must be a list containing the text in the first component.

2. ae.key is a key as described in xyplot.

3. n.events is a vector containing the number of events in each subpanel.

4. panel.widths is a vector of relative widths of the three components of the graph. The numbers must sum to one. Zero values are permitted. The first width includes the left axis and the Percent plot. The second is the Relative Risk plot, and the third is the plot of the table values.

5. AEtable is a table containing the data plotted on its row.

Note

Ann Liu-Ferrara was a beta tester for the shiny app.

Author(s)

Richard M. Heiberger <[email protected]>

References

Ohad Amit, Richard M. Heiberger, and Peter W. Lane. (2008) “Graphical Approaches to the Analysis of Safety Data from Clinical Trials”. Pharmaceutical Statistics, 7, 1, 20–35. https://onlinelibrary.wiley.com/doi/10.1002/pst.254

See Also

AEdotplot.data.frame

Examples

## formula method.  See ?AEdotplot.data.frame for other methods.
data(AEdata)
head(AEdata)

AEdotplot(AE ~ nAE/nTRT, groups = TRT, data = AEdata)           ## sort by Relative Risk
AEdotplot(AE ~ nAE/nTRT | OrgSys, groups = TRT, data = AEdata)  ## conditioned on Organ System

## Not run: 
AEdotplot(AE ~ nAE/nTRT, groups = TRT, data = AEdata, sortbyVar="PCT")                   ## PCT A
AEdotplot(AE ~ nAE/nTRT, groups = TRT, data = AEdata, sortbyVar="PCT", sortbyVarBegin=2) ## PCT B
AEdotplot(AE ~ nAE/nTRT, groups = TRT, data = AEdata, sortbyRelativeRisk=FALSE)     ## levels(AE)
AEdotplot(AE ~ nAE/nTRT | OrgSys, groups = TRT, data = AEdata, sortbyVar="ase.logrelrisk")

## End(Not run)


## Not run: 
AEdotplot(AE ~ nAE/nTRT | OrgSys, groups = TRT,
          data = AEdata[c(AEdata$OrgSys %in% c("GI","Resp")),])

## test sortbyRelativeRisk=FALSE
ABCD.12345 <- AEdata[1:12,]
head(ABCD.12345)
AEdotplot(AE ~ nAE/nTRT | OrgSys, groups=TRT, data=ABCD.12345)
AEdotplot(AE ~ nAE/nTRT | OrgSys, groups=TRT, data=ABCD.12345, sort=FALSE)

## suppress third panel
tmp <- AEdotplot(AE ~ nAE/nTRT, groups = TRT, data = AEdata)
print(tmp, AEtable=FALSE)

## End(Not run)

## Not run: 
  ## run the shiny app
 if (interactive())  shiny::runApp(system.file("shiny/AEdotplot", package="HH"))

## End(Not run)

Description

Support functions for the AEdotplot.

Usage

## S3 method for class 'data.frame'
AEdotplot(xr, ...,
          conditionVariable=NULL,
          conditionName=deparse(substitute(xr)),
          useCondition=!is.null(conditionVariable),
          sub=list(conditionName, cex=.7))

  ## S3 method for class 'AElogrelrisk'
AEdotplot(xr,
          A.name=paste(levels(xr$RAND)[1], " (n=", xr$SN[1], ")", sep=""),
          B.name=paste(levels(xr$RAND)[2], " (n=", xr$SN[2], ")", sep=""),
          col.AB=c("red","blue"), pch.AB=c(16,17),
           main=if (sortbyRelativeRisk)
              list("Most Frequent On-Therapy Adverse Events Sorted by Relative Risk",
                   cex=1)
           else
              list("Most Frequent On-Therapy Adverse Events", cex=1),
          cex.AB.points=NULL, cex.AB.y.scale=.6, cex.x.scale=.6,
          panel.widths=c(.55, .22, .23),
          key.y=-.2, CI.percent=95,
          conditionName=deparse(substitute(xr)),
          sortbyRelativeRisk=TRUE,
          ...,
          sub=list(conditionName, cex=.7),
          par.strip.text=list(cex=.7))

  ## S3 method for class 'AEtable'
AEdotplot(xr, ..., useCondition=TRUE,
                              sub="sub for AEsecond")

  ## S3 method for class 'AEdotplot'
print(x, ...,
      main=attr(x, "main"),
      sub=attr(x,"sub"),
      ae.key=attr(x, "ae.key"),
      panel.widths=attr(x,"panel.widths"),
      AEtable=TRUE)

  ## S3 method for class 'AEdotplot'
c(..., panel.widths=attr(aedp[[1]], "panel.widths"),
                        par.strip.text=list(cex=.7))

  AElogrelrisk(ae,
               A.name=levels(ae$RAND)[1],
               B.name=levels(ae$RAND)[2],
               crit.value=1.96,
               sortbyRelativeRisk=TRUE, ...,
                         sortbyVar=c("PREF", ## Event name
                           "PCT",            ## Percent
                           "SN",             ## Number of Patients
                           "SAE",            ## Number of Observed Events
                           "relrisk",        ## Relative Risk (RR)
                           "ase.logrelrisk", ## Asymptotic Standard Error(log(RR))
                           "relriskCI.lower", ## Confidence Interval Bounds
                           "relriskCI.upper"),
                         sortbyVarBegin=1)   ## 1 for A treatment, 2 for B treatment

  AEmatchSortorder(AEstandard,
                   AEsecond,
                   AEsecond.AEtable=attr(AEsecond, "AEtable"),
                   levels.order=
                      lapply(attr(AEstandard,"AEtable"),
                             function(AEsubtable) levels(AEsubtable$PREF)),
                   main.second=list(paste("Most Frequent On-Therapy Adverse Events",
                                          "Sorted to Match First Table"),
                                    cex=1))

  ## S3 method for class 'AEdotplot'
update(object, ...)

Arguments

.

Details

The first panel is an ordinary dotplot of the percent of AE observed for each treatment by AE.

The second panel shows relative risk of an event on the Treatment B arm (usually the active compound) relative to the Treatment A arm (usually the placebo), with 95% confidence intervals for a $2\times2$ table. Confidence intervals on the log relative risk are calculated using the asymptotic standard error formula given as Equation 3.18 in Agresti A., Categorical Data Analysis. Wiley: New York, 1990.

By default the AEdotplot function sorts the events by relative risk. To retain the sort order implied by the levels of the AE factor, specify the argument sortbyRelativeRisk=FALSE. To control the sort order, make the AE factor in the input dataset an ordered factor and specify the levels in the order you want.

The third panel shows the numerical values of the number and percent of observed events on each arm and the relative risk. The display of third panel can be suppressed by specifying the panel.widths argument.

Value

The primary interest is in the display of the plot.

The function returns an AEdotplot object which is a list of three trellis objects, one for the the Percent plot, one for the Relative Risk plot, and one for the Text plot containing the table of input values. The object has attributes

1. main and sub hold the main and subtitles. Each must be a list containing the text in the first component.

2. ae.key is a key as described in xyplot.

3. n.events is a vector containing the number of events in each subpanel.

4. panel.widths is a vector of relative widths of the three components of the graph. The numbers must sum to one. Zero values are permitted. The first width includes the left axis and the Percent plot. The second is the Relative Risk plot, and the third is the plot of the table values.

5. AEtable is a table containing the data plotted on its row.

Author(s)

Richard M. Heiberger <[email protected]>

References

Ohad Amit, Richard M. Heiberger, and Peter W. Lane. (2008) “Graphical Approaches to the Analysis of Safety Data from Clinical Trials”. Pharmaceutical Statistics, 7, 1, 20–35. https://onlinelibrary.wiley.com/doi/10.1002/pst.254

See Also

AEdotplot

Examples

## Not run: 
## variable names in the input data.frame aeanonym
## RAND   treatment as randomized
## PREF   adverse event symptom name
## SN     number of patients in treatment group
## SAE    number of patients  in each group for whom the event PREF was observed
## OrgSys Organ System
##
## Input sort order is PREF/RAND

data(aeanonym)
head(aeanonym)

## variable names are hard-wired in the program
## names(aeanonym) <- c("RAND", "PREF", "SAE", "SN", "OrgSys")


## Calculate log relative risk and confidence intervals (95
## AElogrelrisk sets the sort order for PREF to match the relative risk.
aeanonymr <- AElogrelrisk(aeanonym) ## PREF sorted by relative risk
head(aeanonymr)
class(aeanonymr$PREF)
levels(aeanonymr$PREF)

AEdotplot(aeanonym)

AEdotplot(aeanonym, sort=FALSE)

AEdotplot(aeanonym, conditionVariable=aeanonym$OrgSys)

aefake <- rbind(cbind(aeanonym, group="ABC"), cbind(aeanonym, group="DEF"))
aefake$SAE[67:132] <- sample(aefake$SAE[67:132])
aefake$OrgSys.group <- with(aefake, interaction(OrgSys, group))

## fake 2
KEEP <- aefake$OrgSys %in% c("GI","Resp")
AEfakeGR <- AEdotplot(aefake[KEEP,], conditionVariable=aefake$OrgSys.group[KEEP],
            sub=list("ABC and DEF have different sort orders for PREF", cex=.7))
AEfakeGR ## ABC and DEF have different sort orders for PREF

AEfakeGR1 <- AEdotplot(aefake[KEEP & (1:132) <= 66,],
                       conditionVariable=aefake$OrgSys.group[KEEP & (1:132) <= 66])
AEfakeGR2 <- AEdotplot(aefake[KEEP & (1:132) >= 67,],
                       conditionVariable=aefake$OrgSys.group[KEEP & (1:132) >= 67])

AEfakeGR1
AEfakeGR2

AEfakeMatched <- AEmatchSortorder(AEfakeGR1, AEfakeGR2)
update(do.call(c, AEfakeMatched),
       main="ABC sorted by Relative Risk; DEF matches ABC order")

## End(Not run)
## Please see  ?AEdotplot  for examples using the formula method
##
## Many more examples are in demo("AEdotplotManyExamples")

Description

Compute and plot oneway analysis of covariance. The result object is an ancova object which consists of an ordinary aov object with an additional trellis attribute. The trellis attribute is a trellis object consisting of a series of plots of y ~ x. The left set of panels is conditioned on the levels of the factor groups. The right panel is a superpose of all the groups.

Usage

ancova(formula, data.in = NULL, ...,
       x, groups, transpose = FALSE,
       display.plot.command = FALSE,
       superpose.level.name = "superpose",
       ignore.groups = FALSE, ignore.groups.name = "ignore.groups",
       blocks, blocks.pch = letters[seq(levels(blocks))],
       layout, between, main,
       pch=trellis.par.get()$superpose.symbol$pch)

panel.ancova(x, y, subscripts, groups,
 transpose = FALSE, ...,
 coef, contrasts, classes,
 ignore.groups, blocks, blocks.pch, blocks.cex, pch)

## The following are ancova methods for generic functions.
## S3 method for class 'ancova'
anova(object, ...)

## S3 method for class 'ancova'
predict(object, ...)

## S3 method for class 'ancova'
print(x, ...) ## prints the anova(x) and the trellis attribute

## S3 method for class 'ancova'
model.frame(formula, ...)

## S3 method for class 'ancova'
summary(object, ...)

## S3 method for class 'ancova'
plot(x, y, ...) ## standard lm plot.  y is always ignored.

## S3 method for class 'ancova'
coef(object, ...)

Arguments

object. The functions using this argument are methods for the similarly named generic functions.

Details

The ancova function does two things. It passes its arguments directly to the aov function and returns the entire aov object. It also rearranges the data and formula in its argument and passes that to the xyplot function. The trellis attribute is a trellis object consisting of a series of plots of y ~ x. The left set of panels is conditioned on the levels of the factor groups. The right panel is a superpose of all the groups.

Value

The result object is an ancova object which consists of an ordinary aov object with an additional trellis attribute. The default print method is to print both the anova of the object and the trellis attribute.

Author(s)

Richard M. Heiberger <[email protected]>

References

Heiberger, Richard M. and Holland, Burt (2015). Statistical Analysis and Data Display: An Intermediate Course with Examples in R. Second Edition. Springer-Verlag, New York. https://link.springer.com/book/10.1007/978-1-4939-2122-5

See Also

ancova-class aov xyplot. See ancovaplot for a newer set of functions that keep the graph and the aov object separate.

Examples

data(hotdog)

## y ~ x                     ## constant line across all groups
ancova(Sodium ~ Calories,     data=hotdog, groups=Type)

## y ~ a                     ## different horizontal line in each group
ancova(Sodium ~            Type, data=hotdog, x=Calories)

## This is the usual usage
## y ~ x + a  or  y ~ a + x  ## constant slope, different intercepts
ancova(Sodium ~ Calories + Type, data=hotdog)
ancova(Sodium ~ Type + Calories, data=hotdog)

## y ~ x * a  or  y ~ a * x  ## different slopes, and different intercepts
ancova(Sodium ~ Calories * Type, data=hotdog)
ancova(Sodium ~ Type * Calories, data=hotdog)

## y ~ a * x ## save the object and print the trellis graph
hotdog.ancova <- ancova(Sodium ~ Type * Calories, data=hotdog)
attr(hotdog.ancova, "trellis")


## label points in the panels by the value of the block factor
data(apple)
ancova(yield ~ treat + pre, data=apple, blocks=block)

## Please see
##      demo("ancova")
## for a composite graph illustrating the four models listed above.

Description

Analysis of Covariance. The class is an extension of "aov" and "lm". It is identical to the "aov" for a single factor and a single covariate plus an attribute which contains a "trellis" object. Four different models are included in the class. See ancova for the examples.

Objects from the Class

A virtual Class: No objects may be created from it.

Extends

Class "aov", directly. Class "lm", by class "aov", distance 2. Class "mlm", by class "aov", distance 2, with explicit test and coerce. Class "oldClass", by class "aov", distance 3. Class "oldClass", by class "aov", distance 4, with explicit test and coerce.

Methods

No methods defined with class "ancova" in the signature. S3-type methods are "anova.ancova", "coef.ancova", "coefficients.ancova", "model.frame.ancova", "plot.ancova", "predict.ancova", "print.ancova", "summary.ancova". "plot.ancova(x)" plots a standard lm plot of x. "print.ancova(x)" prints the anova(x) and the trellis attribute. The remaining methods use NextMethod.

Author(s)

Richard M. Heiberger <[email protected]>

References

Heiberger, Richard M. and Holland, Burt (2015). Statistical Analysis and Data Display: An Intermediate Course with Examples in R. Second Edition. Springer-Verlag, New York. https://link.springer.com/book/10.1007/978-1-4939-2122-5

See Also

ancova

Description

Analysis of Covariance Plots. Any of the ancova models
y ~ x * t
y ~ t * x
y ~ x + t
y ~ t + x
y ~ x , groups=t
y ~ t, x=x
y ~ x * t, groups=b
y ~ t * x, groups=b
y ~ x + t, groups=b
y ~ t + x, groups=b

Usage

ancovaplot(object, ...)
## S3 method for class 'formula'
ancovaplot(object, data, groups=NULL, x=NULL, ...,
           formula=object,
           col=rep(tpg$col,
             length=length(levels(as.factor(groups)))),
           pch=rep(c(15,19,17,18,16,20, 0:14),
             length=length(levels(as.factor(groups)))),
           slope, intercept,
           layout=c(length(levels(cc)), 1),
           col.line=col, lty=1,
           superpose.panel=TRUE,
           between=if (superpose.panel)
                      list(x=c(rep(0, length(levels(cc))-1), 1))
                   else
                      list(x=0),
           col.by.groups=FALSE ## ignored unless groups= is specified
           )

panel.ancova.superpose(x, y, subscripts, groups,
                       slope, intercept,
                       col, pch, ...,
                       col.line, lty,
                       superpose.panel,
                       col.by.groups,
                       condition.factor,
                       groups.cc.incompatible,
                       plot.resids=FALSE,
                       print.resids=FALSE,
                       mean.x.line=FALSE,
                       col.mean.x.line="gray80")

Arguments

Details

Value

ancovaplot returns a c("ancova","trellis") object. panel.ancova.superpose is an ordinary lattice panel function.

Author(s)

Richard M. Heiberger <[email protected]>

References

Heiberger, Richard M. and Holland, Burt (2015). Statistical Analysis and Data Display: An Intermediate Course with Examples in R. Second Edition. Springer-Verlag, New York. https://link.springer.com/book/10.1007/978-1-4939-2122-5

See Also

See the older ancova.

Examples

data(hotdog, package="HH")
  ancovaplot(Sodium ~ Calories + Type, data=hotdog)
  ancovaplot(Sodium ~ Calories * Type, data=hotdog)
  ancovaplot(Sodium ~ Calories, groups=Type, data=hotdog)
  ancovaplot(Sodium ~ Type, x=Calories, data=hotdog)

  ## Please see demo("ancova", package="HH") to coordinate placement
  ## of all four of these plots on the same page.

  ancovaplot(Sodium ~ Calories + Type, data=hotdog, plot.resids=TRUE)

Description

ANOVA table for a c("ancovaplot","trellis") object.

Usage

## S3 method for class 'ancovaplot'
anova(object, ...)
aov.ancovaplot(object, warn=TRUE)
aovStatement(object, ...)
## S3 method for class 'ancovaplot'
aovStatement(object, ...)
aovStatementAndAnova(object, ...)
## S3 method for class 'ancovaplot'
aovStatementAndAnova(object, ...)
## S3 method for class 'ancovaplot'
model.tables(x, ...)

Arguments

Details

The aov.ancovaplot modifies the call item into an aov call with the same formula and data. If there are groups in the call specified as a name, the groups factor is included in the constructed aov call only if there are both a factor and a covariate in the right-hand-side of the formula. In that case they the groups will be interpreted as a block factor and will be placed first. If the groups are specified as a vector of values in the call, the groups are ignored with a warning. If there is only one term in the right-hand-side, then the groups factor will not be placed into the aov formula. In this case, there will be a warning if the argument warn is TRUE, and no warning if the warn argument is FALSE.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

ancovaplot

Description

Oneway ANOVA table from the summary information consisting of group sample sizes, means, and standard deviations. The full dataset is not needed.

Usage

anovaMean(object, n, ybar, s, ..., ylabel = "ylabel")

Arguments

Value

Analysis of variance table, identical to the ANOVA table that would have been produced by anova.lm if the original data, rather than the summary data, had been available.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

anova.lm,

plot.mmc.multicomp

Examples

## pulmonary data used in Hsu and Peruggia paper defining the mean-mean plot
## See ?plot.mmc.multicomp for details on the dataset.

data(pulmonary)

anovaMean(pulmonary$smoker,
          pulmonary$n,
          pulmonary$FVC,
          pulmonary$s,
          ylabel="pulmonary")

Description

Analysis of variance from sufficient statistics for groups. For each group, we need the factor level, the response mean, the within-group standard deviation, and the sample size. The correct ANOVA table is produced. The residuals are fake. The generic vcov and summary.lm don't work for the variance of the regression coefficients in this case. Use vcovSufficient.

Usage

aovSufficient(formula, data = NULL,
              projections = FALSE, qr = TRUE, contrasts = NULL,
              weights = data$n, sd = data$s,
              ...)

vcovSufficient(object, ...)

Arguments

Value

For aovSufficient, an object of class c("aov", "lm"). For vcovSufficient, a function that returns the covariance matrix of the regression coefficients.

Note

The residuals are fake. They are all identical and equal to the MLE standard error (sqrt(SumSq.res/df.tot)). They give the right ANOVA table. They may cause confusion or warnings in other programs. The standard errors and t-tests of the coefficients are not calculated by summary.lm. Using the aov object from aovSufficient in glht requires the vcov. and df arguments.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

MMC and aov.

Examples

## This example is from Hsu and Peruggia

## This is the R version
## See ?mmc.mean for S-Plus

if.R(s={},
r={

data(pulmonary)
pulmonary
pulmonary.aov <- aovSufficient(FVC ~ smoker,
                                data=pulmonary)
summary(pulmonary.aov)

## Not run: 
pulmonary.mmc <- mmc(pulmonary.aov,
                     linfct=mcp(smoker="Tukey"),
                     df=pulmonary.aov$df.residual,
                     vcov.=vcovSufficient)
mmcplot(pulmonary.mmc, style="both")

## orthogonal contrasts
pulm.lmat <- cbind("npnl-mh"=c( 1, 1, 1, 1,-2,-2), ## not.much vs lots
                   "n-pnl"  =c( 3,-1,-1,-1, 0, 0), ## none vs light
                   "p-nl"   =c( 0, 2,-1,-1, 0, 0), ## {} arbitrary 2 df
                   "n-l"    =c( 0, 0, 1,-1, 0, 0), ## {} for 3 types of light
                   "m-h"    =c( 0, 0, 0, 0, 1,-1)) ## moderate vs heavy
dimnames(pulm.lmat)[[1]] <- row.names(pulmonary)
pulm.lmat

pulmonary.mmc <- mmc(pulmonary.aov,
                     linfct=mcp(smoker="Tukey"),
                     df=pulmonary.aov$df.residual,
                     vcov.=vcovSufficient,
                     focus.lmat=pulm.lmat)

mmcplot(pulmonary.mmc, style="both", type="lmat")

## End(Not run)
})

Description

Repair design error in S-Plus arima.diag.

Usage

arima.diag.hh(z, acf.resid = TRUE,
              lag.max = round(max(gof.lag + n.parms + 1, 10 * log10(n))),
              gof.lag = 15, resid = FALSE,
              std.resid = TRUE, plot = TRUE, type = "h", ...,
              x=eval(parse(text = series.name)))

Arguments

Details

Repairs design flaw in S-Plus arima.diag. The location of the time series is hardwired one level up, so it can't be found when arima.diag is not one level down from the top.

This function is a no-op in R.

Value

This function is a no-op in R. It returns NA.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

tsdiagplot in both systems and arima.diag

in S-Plus.

Description

Loop through a series of ARIMA models and display coordinated tables and diagnostic graphs. The complete example from the Heiberger and Teles article, also included in the Heiberger and Holland book, is illustrated.

Usage

arma.loop(x,
          model,             ## S-Plus
          order, seasonal,   ## R
          series=deparse(substitute(x)), ...)

diag.arma.loop(z,
               x=stop("The time series x is needed in S-Plus when p=q=0."),
               lag.max = 36, gof.lag = lag.max)

rearrange.diag.arma.loop(z)

Arguments

Details

S-Plus and R have different functions, with different input argument names and different components in their value.

Value

arma.loop: "arma.loop" object which is a matrix of lists, each containing an arima model.

diag.arma.loop: "diag.arma.loop" object which is a matrix of lists, each containing the standard diagnostics for one arima model.

rearrange.diag.arma.loop: List of matrices, each containing all the values for a specific diagnostic measure collected from the set of arima models.

Author(s)

Richard M. Heiberger ([email protected])

References

"Displays for Direct Comparison of ARIMA Models" The American Statistician, May 2002, Vol. 56, No. 2, pp. 131-138. Richard M. Heiberger, Temple University, and Paulo Teles, Faculdade de Economia do Porto.

Heiberger, Richard M. and Holland, Burt (2015). Statistical Analysis and Data Display: An Intermediate Course with Examples in R. Second Edition. Springer-Verlag, New York. https://link.springer.com/book/10.1007/978-1-4939-2122-5

See Also

tsdiagplot

Examples

## see tsdiagplot for the example

Description

Constructs class="likert" objects to be used by the plot.likert methods.

Usage

is.likert(x)

as.likert(x, ...)
## Default S3 method:
as.likert(x, ...)
## S3 method for class 'data.frame'
as.likert(x, ...)
## S3 method for class 'formula'
as.likert(x, ...)  ## doesn't work yet
## S3 method for class 'ftable'
as.likert(x, ...)
## S3 method for class 'table'
as.likert(x, ...)
## S3 method for class 'matrix'
as.likert(x,
          ReferenceZero=NULL,
          ...,
          rowlabel=NULL, collabel=NULL,
          xlimEqualLeftRight=FALSE,
          xTickLabelsPositive=TRUE,
          padding=FALSE,
          reverse.left=TRUE)
## S3 method for class 'listOfNamedMatrices'
as.likert(x, ...)
## S3 method for class 'array'
as.likert(x, ...)

## S3 method for class 'likert'
rev(x)

is.likertCapable(x, ...)

Arguments

Details

Please see likert for information on the plot for which as.likert prepares the data.

Value

For the as.likert methods, a likert object, which is a matrix with additional attributes that are needed to make the barchart method used by the plot.likert methods work with the data. Columns for respondents who disagree have negated values. Any NA values in the argument x are changed to 0. The column of the original data for respondents who neither agree nor disagree is split into two columns, each containing halved values—one positive and one negative. Negative columns come first in the sequence of "No Opinion"(negative)–"Strongly Disagree", followed by "No Opinion"(positive)–"Strongly Agree". There are four attributes: "even.col" indicating whether there were originally an even number of columns, "n.levels" the original number of levels, "levels" the original levels in the original order, "positive.order" The sequence in which to display the rows in order to make the right hand sides progress with high values on top.

is.likert returns a TRUE or FALSE value.

is.likertCapable returns a TRUE or FALSE value if the argument can used as an argument to one of the plot.likert methods.

Author(s)

Richard M. Heiberger <[email protected]>

References

Richard M. Heiberger, Naomi B. Robbins (2014)., "Design of Diverging Stacked Bar Charts for Likert Scales and Other Applications", Journal of Statistical Software, 57(5), 1–32, doi:10.18637/jss.v057.i05.

Naomi Robbins <[email protected]>, "Visualizing Data: Challenges to Presentation of Quality Graphics—and Solutions", Amstat News, September 2011, 28–30.

Naomi B. Robbins and Richard M. Heiberger (2011). Plotting Likert and Other Rating Scales. In JSM Proceedings, Section on Survey Research Methods. Alexandria, VA: American Statistical Association.

Luo, Amy and Tim Keyes (2005). "Second Set of Results in from the Career Track Member Survey," Amstat News. Arlington, VA: American Statistical Association.

See Also

likert

Examples

## Please see ?likert to see these functions used in context.

tmp2 <- array(1:12, dim=c(3,4), dimnames=list(B=LETTERS[3:5], C=letters[6:9]))
as.likert(tmp2)  ## even number of levels.

is.likert(tmp2)
is.likert(as.likert(tmp2))

Description

Convert a list of numeric matrices to a single matrix. This function is used to improve legibility of the printed object. The as.matrix.listOfNamedMatrices display is easier to read when the rownames are very long, as in the example illustrated here. Because the default print of the matrix repeats the rownames several times, with only a few columns of the data shown in each repetition, the actual matrix structure of the data values is obscured.

Usage

## S3 method for class 'listOfNamedMatrices'
as.matrix(x, abbreviate = TRUE, minlength = 4, ...)
is.listOfNamedMatrices(x,  xName=deparse(substitute(x)))
## S3 method for class 'listOfNamedMatrices'
as.data.frame(x, ...)
as.listOfNamedMatrices(x,  xName=deparse(substitute(x)), ...)
## S3 method for class 'listOfNamedMatrices'
x[...]
## S3 method for class 'array'
as.listOfNamedMatrices(x, xName=deparse(substitute(x)), ...)
## S3 method for class 'list'
as.listOfNamedMatrices(x, xName=deparse(substitute(x)), ...)
## S3 method for class 'MatrixList'
as.listOfNamedMatrices(x, xName=deparse(substitute(x)), ...)
## S3 method for class 'listOfNamedMatrices'
print(x, ...)

as.MatrixList(x)
## S3 method for class 'array'
as.MatrixList(x)
## S3 method for class 'MatrixList'
print(x, ...)

as.likertDataFrame(x, xName=deparse(substitute(x)))
## S3 method for class 'listOfNamedMatrices'
as.likertDataFrame(x, xName=deparse(substitute(x)))
## S3 method for class 'array'
as.likertDataFrame(x, xName=deparse(substitute(x)))

Arguments

Value

The result of as.listOfNamedMatrices is a list with class=c("listOfNamedMatrices", "list").

The result of as.matrix.listOfNamedMatrices is an rbind of the individual matrices in the argument list x. The rownames of the result matrix are constructed by pasting the abbreviation of the list item names with the abbreviation of the individual matrix rownames. The original names are retained as the "Subtables.Rows" attribute.

The result of is.listOfNamedMatrices is logical value.

print.listOfNamedMatrices prints as.matrix.listOfNamedMatrices of its argument and returns the original argument.

as.data.frame.listOfNamedMatrices(x, ...) is an unfortunate kluge. The result is the original x that has NOT been transformed to a data.frame. A warning message is generated that states that the conversion has not taken place. This kluge is needed to use "listOfNamedMatrices" objects with the Commander package because Rcmdr follows its calls to the R data function with an attempt, futile in this case, to force the resulting object to be a data.frame.

The as.MatrixList methods construct a list of matrices from an array. Each matrix has the first two dimensions of the array. The result list is itself an array defined by all but the first two dimensions of the argument array.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

likert

Examples

data(ProfChal)

tmp <- data.matrix(ProfChal[,1:5])
rownames(tmp) <- ProfChal$Question
ProfChal.list <- split.data.frame(tmp, ProfChal$Subtable)

## Original list of matrices is difficult to read because
## it is displayed on too many lines.
ProfChal.list[2:3]

## Single matrix with long list item names and long row names
## of argument list retained as an attribute.
as.listOfNamedMatrices(ProfChal.list[2:3], minlength=6)

## Not run: 
## NA as a dimname value
tmp <- structure(c(0, 0, 0, 6293, 18200, 2122,
                   0, 0, 0, 2462, 7015, 5589, 
                   6908, 5337, 842, 0, 0, 0),
                 .Dim = c(3L, 2L, 3L),
                 .Dimnames = list(c("A", "B", "C"),
                                  c("D", "E"),
                                  c("F", "G", NA)))
tmp
as.MatrixList(tmp)

## End(Not run)

## Not run: 
sapply(as.MatrixList(tmp3), as.likert, simplify=FALSE)  ## odd number of levels.

data(NZScienceTeaching)
likert(Question ~ ., NZScienceTeaching)
likert(Question ~ . | Subtable, data=NZScienceTeaching)
likert(Question ~ . | Subtable, data=NZScienceTeaching,
       layout=c(1,2), scales=list(y=list(relation="free")))



## End(Not run)

Description

MMC plots: In R, functions used to interface the glht in R to the MMC functions designed with S-Plus multicomp notation. These are all internal functions that the user doesn't see.

Usage

## S3 method for class 'mmc.multicomp'
print(x, ..., width.cutoff=options()$width-5)

## S3 method for class 'multicomp'
print(x, ...)

## print.multicomp.hh(x, digits = 4, ..., height=T) ## S-Plus only

## S3 method for class 'multicomp.hh'
print(x, ...) ## R only


as.multicomp(x, ...)

## S3 method for class 'glht'
as.multicomp(x,       ## glht object
           focus=x$focus,
           ylabel=deparse(terms(x$model)[[2]]),
           means=model.tables(x$model, type="means",
                              cterm=focus)$tables[[focus]],
           height=rev(1:nrow(x$linfct)),
           lmat=t(x$linfct),
           lmat.rows=lmatRows(x, focus),
           lmat.scale.abs2=TRUE,
           estimate.sign=1,
           order.contrasts=TRUE,
           contrasts.none=FALSE,
           level=0.95,
           calpha=NULL,
           method=x$type,
           df,
           vcov.,
           ...
           )

as.glht(x, ...)

## S3 method for class 'multicomp'
as.glht(x, ...)

Arguments

Details

The mmc.multicomp print method displays the confidence intervals and heights on the MMC plot for each component of the mmc.multicomp object.

print.multicomp displays the confidence intervals and heights for a single component.

Value

as.multicomp is a generic function to change its argument to a "multicomp" object.

as.multicomp.glht changes an "glht" object to a "multicomp" object. If the model component of the argument "x" is an "aov" object then the standard error is taken from the anova(x$model) table, otherwise from the summary(x). With a large number of levels for the focus factor, the summary(x) function is exceedingly slow (80 minutes for 30 levels on 1.5GHz Windows XP). For the same example, the anova(x$model) takes a fraction of a second.

Note

The multiple comparisons calculations in R and S-Plus use completely different functions. MMC plots in R are based on glht.

MMC plots in S-Plus are based on multicomp.

The MMC plot is the same in both systems. The details of gettting the plot differ.

Author(s)

Richard M. Heiberger <[email protected]>

References

Heiberger, Richard M. and Holland, Burt (2015). Statistical Analysis and Data Display: An Intermediate Course with Examples in R. Second Edition. Springer-Verlag, New York. https://link.springer.com/book/10.1007/978-1-4939-2122-5

Heiberger, Richard M. and Holland, Burt (2006). "Mean–mean multiple comparison displays for families of linear contrasts." Journal of Computational and Graphical Statistics, 15:937–955.

See Also

mmc, glht.

Description

as.vector.trellis converts a two-dimensional trellis object into a one-dimensional trellis object. reorder.trellis changes the order of the panel.args component in a trellis object. These are are mostly used as utilities by matrix.trellis.

Usage

## S3 method for class 'trellis'
as.vector(x, mode = "any")
## S3 method for class 'trellis'
reorder(x, X, ...)

Arguments

Value

trellis object with length(dim(x)) == 1. as.vector retains the original order of the panels. reorder changes the order to the one specified by using the X argument as a subscript.

Author(s)

Richard M. Heiberger <[email protected]>

Examples

tmp <- data.frame(a=letters[c(1:3,1:3,1:3)],
                  b=1:9,
                  d=1:9,
                  e=LETTERS[c(4,4,4,5,5,5,6,6,6)])
tmp
a6 <- xyplot(b ~ d | a*e, data=tmp, pch=19)
a6
dim(a6)
a62 <- as.vector(a6)
a62
dim(a62)
a63 <- reorder(a6, c(1,4,7, 2,5,8, 3,6,9))
a63
dim(a63)
a64 <- matrix.trellis(a63, nrow=3, ncol=3, dimnames=dimnames(a6), byrow=TRUE)
a64
dim(a64)

Description

Labels the bottom axis with the x-factor name for each column. Labels the right axis with the response variable name in all rows.

Usage

axis.i2wt(side, scales, ...)

Arguments

Author(s)

Richard M. Heiberger, with asssistance from Deepayan Sarkar.

See Also

interaction2wt

Description

Plot the bivariate normal density using wireframe for specified rho. There is a shiny app that allows this to be done dynamically.

Usage

bivariateNormal(rho = 0, layout = c(3, 3), lwd = 0.2,
                angle = c(22.5, 67.5, 112.5, 337.5, 157.5, 292.5, 247.5, 202.5),
                col.regions = trellis.par.get("regions")$col, ...)

Arguments

Details

The default setting shows the view as seen from a series of eight angles. To see just a single view, see the example.

Value

"trellis" object.

Note

Based on the galaxy example on pages 204–205 in S & S-PLUS Trellis Graphics User's Manual, Richard A. Becker and William S. Cleveland (1996), https://www.stat.auckland.ac.nz/~ihaka/courses/120/trellis.user.pdf

Author(s)

Richard M. Heiberger ([email protected])

Examples

bv8 <- bivariateNormal(.7)  ## all views on one page
bv8
update(bv8[3], layout=c(1,1)) ## one panel
## Not run: 
  if (interactive())
    shiny::runApp(file.path(system.file(package="HH"), "shiny/bivariateNormal")) ## 3D
  if (interactive())
    shiny::runApp(system.file("shiny/bivariateNormalScatterplot", package="HH")) ## scatterplot

## End(Not run)

Description

The data, the least squares line, the confidence interval lines, and the prediction interval lines for a simple linear regression (lm(y ~ x)) are displayed. Tick marks are placed at the location of xbar, the x-value of the narrowest interval.

Usage

ci.plot(lm.object, ...)

## S3 method for class 'lm'
ci.plot(lm.object,
        xlim=range(data[, x.name]),
        newdata,
        conf.level=.95,
        data=model.frame(lm.object),
        newfit,
        ylim,
        pch=19,
        lty=c(1,3,4,2),
        lwd=2,
        main.cex=1,
        main=list(paste(100*conf.level,
          "% confidence and prediction intervals for ",
          substitute(lm.object), sep=""), cex=main.cex), ...
        )

Arguments

Value

"trellis" object containing the plot.

Note

The predict.lm functions in S-Plus and R differ. The S-Plus function can produce both confidence and prediction intervals with a single call. The R function produces only one of them in a single call. Therefore the default calculation of newfit within the function depends on the system.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

lm, predict.lm

Examples

tmp <- data.frame(x=rnorm(20), y=rnorm(20))
tmp.lm <- lm(y ~ x, data=tmp)
ci.plot(tmp.lm)

Description

Illustration of the meaning of confidence levels. Generate sets of confidence intervals for independent randomly generated sets of normally distributed numbers. Low confidence levels give narrow intervals that are less likely to bracket the true value. Higher confidence levels increase the probability of bracketing the true value, and are also much wider and therefore less precise. The shiny app can animate how the increase in confidence level and width leads to a consequent decrease in precision.

Usage

CIplot(n.intervals = 100,
       n.per.row = 40,
       pop.mean = 0,
       pop.sd = 1,
       conf.level = 0.95,
       ...)

confintervaldata(n.intervals = 100,
                 n.per.row = 40,
                 pop.mean = 0,
                 pop.sd = 1,
                 conf.level = 0.95,
                 seed,
                 ...)

confinterval.matrix(x,
                    conf.level = attr(x, "conf.level"),
                    ...)

confintervalplot(x.ci,
                 n.intervals = nrow(x.ci),
                 pop.mean = attr(x.ci, "pop.mean"),
                 pop.sd = attr(x.ci, "pop.sd"),
                 n.per.row = attr(x.ci, "n.per.row"),
                 xlim, ylim, ...)

shiny.CIplot(height = "auto")

Arguments

Details

The shiny app has sliders for the n.intervals, n.per.row, pop.mean, pop.sd, and conf.level. Changes in the conf.level slider, either manually by animation, use the same set of generated data to show how increasing the confidence level increases the width of the confidence interval and consequently decreases the precision of the interval estimator.

Value

CIplot and confintervalplot return a "trellis" plot containing a plot of Confidence Intervals.

confintervaldata returns a matrix of n.intervals rows by n.per.row columns of independent normally distributed random numbers. The matrix has a set of attributes recording the arguments to the function.

confinterval.matrix returns a data.frame of n.intervals with three columns containing the lower bound, center, and upper bound of the confidence interval for each row of its input matrix. The data.frame has a set of attributes recording the arguments to the function.

shiny.CIplot returns a shiny app object which, when printed, runs a shiny app displaying the Confidence Interval plot and several slider controls.

Author(s)

Richard M. Heiberger <[email protected]>

Examples

## A. from the console

## example 1
CIplot()

## example 2
## Not run: 
CIplot(n.intervals=100,
       n.per.row=40,
       pop.mean=0,
       pop.sd=1,
       conf.level=.95)

## End(Not run)

## example 3
## Not run: 
tmp.data <- confintervaldata()
tmp.ci <- confinterval.matrix(tmp.data)
confintervalplot(tmp.ci)

## End(Not run)

## example 4
## Not run: 
tmp.data <- confintervaldata(n.intervals=100,
                             n.per.row=40,
                             pop.mean=0,
                             pop.sd=1,
                             conf.level=.95)
tmp.ci <- confinterval.matrix(tmp.data)
confintervalplot(tmp.ci)

## End(Not run)


## B. shiny, initiated from the console

## example 5
## Not run: 
  if (interactive())
    shiny.CIplot()

## End(Not run)

## example 6
## Not run: 
  if (interactive())
    shiny.CIplot(height=800)  ## px
  ## take control of the height of the graph in the web page

## End(Not run)

Description

Initialization of an R display device with the graphical parameters that rmh prefers.

Usage

col.hh()

Value

List of graphical parameters to be used in the theme argument to the trellis.device or trellis.par.set functions.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

trellis.device, trellis.par.get

Examples

## Not run: 
trellis.device(theme="col.hh") ## Open a device with the theme

trellis.device(theme=col.hh()) ## Open a device with the theme

trellis.par.set(theme=col.hh())## Add theme to already open device

## End(Not run)

Description

col3x2 color dataset.

Usage

data("col3x2")

Format

The format is: chr [1:6] "#1B9E77" "#D95F02" "#7570B3" "#66C2A5" "#FC8D62" "#8DA0CB"

Details

3x2 color scheme. These colors look like a 3x2 color array when run through the vischeck simulator to see how they look for the three most common color vision deficiencies: Deuteranope, Protanope, Tritanope.

References

About 10% of the population have color deficient vision. Your job is make your graphs legible to everyone. Download ImageJ from https://imagej.net/Downloads and VischeckJ from http://vischeck.com and follow the instructions in those sites. This program will allow you to simulate color deficient vision on your computer.

On my Mac, I need to doubleclick ij.jar to open the program. Then open the "Vischeck Panel" on the Plugins menu and navigate to a png file with the "File Open" menu. Click on each of the three types of color deficiency.

Examples

data(col3x2)
col3x2

## Not run: 
library(RColorBrewer)
library(lattice)
col3x2 <- c(brewer.pal(n=3, "Dark2"), brewer.pal(n=3, "Set2"))
col3x2
## save(col3x2, file="col3x2.rda") ## data(col3x2, package="HH")

## End(Not run)

## Not run: 
barchart(~ 1:6, col=col3x2, lwd=0, origin=0, horizontal=FALSE,
         scales=list(x=list(at=1:6, labels=col3x2)))


tmp <- array(c(1, rep(0,6)), c(1,3,2,6),
             dimnames=list("",
                           c("green","orange","blue"),
                           c("Dark2","Set2"),
                           col3x2))

useOuterStrips(barchart(tmp, col=col3x2,
                        between=list(x=1, y=1),
                        scales=list(x=list(at=-1)),
                        main="col3x2", xlab="")) +
  layer(panel.text(x=.5, y=1.45, labels=col3x2[panel.number()]))


## End(Not run)

Description

Combine limits on a one-dimensional trellis object.

Usage

combineLimits.trellisvector(x, margin.x = 1:2, margin.y = 1:2,
                            layout = x$layout,
                            ncol=x$layout[1], nrow=x$layout[2],
                            condlevels = x$condlevels[[1]], ...)

Arguments

Details

The one-dimensional object is converted to a two-dimensional object which is forwarded to the standard combineLimits function. The result is converted back to a one-dimensional object.

Value

One-dimensional trellis object with combined xlim and ylim values across all panels.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

combineLimits

Examples

tmp <- data.frame(a=1:3,
                  b=c(4,5,7),
                  c=5:7,
                  d=c(8, 9, 12),
                  e=9:11)
tmp

a2 <- xyplot(a + b ~ c + d + e, data=tmp, outer=TRUE,
             scales=list(relation="free"), main="a2")
a2
dim(a2)
combineLimits.trellisvector(a2)
a21 <- combineLimits.trellisvector(update(a2, layout=c(3,2)))
a21
dim(a21)

Description

Rearranges and improves the legibility of the output from the stepwise

function in S-Plus. The output can be used for the Cp plot. cp.calc works only in S-Plus. Use regsubsets

in R. The example below works in both languages.

Usage

cp.calc(sw, data, y.name)

## S3 method for class 'cp.object'
print(x, ...)

## S3 method for class 'cp.object'
x[..., drop = TRUE]

Arguments

Value

"cp.object", which is a data.frame containing information about each model that was attempted with additional attributes: tss total sum of squares, n number of observations, y.name response variable, full.i row name of full model. The columns are

Author(s)

Richard M. Heiberger <[email protected]>

References

Heiberger, Richard M. and Holland, Burt (2015). Statistical Analysis and Data Display: An Intermediate Course with Examples in R. Second Edition. Springer-Verlag, New York. https://link.springer.com/book/10.1007/978-1-4939-2122-5

Examples

## This example is from Section 9.15 of Heiberger and Holland (2004).
data(usair)
if.R(s={usair <- usair}, r={})

splom(~usair, main="U.S. Air Pollution Data with SO2 response", cex=.5)
## export.eps(hh("regb/figure/regb.f1.usair.eps"))

usair$lnSO2 <- log(usair$SO2)
usair$lnmfg <- log(usair$mfgfirms)
usair$lnpopn <- log(usair$popn)

usair[1:3,]   ## lnSO2 is in position 8, SO2 is in position 1
              ## lnmfg is in position 9, lnpopn is in position 10

splom(~usair[, c(8,2,9,10,5:7)],
              main="U.S. Air Pollution Data with 3 log-transformed variables",
              cex=.5)
## export.eps(hh("regb/figure/regb.f2.usair.eps"))

if.R(s={
  usair.step <- stepwise(y=usair$lnSO2,
                         x=usair[, c(2,9,10,5:7)],
                         method="exhaustive",
                         plot=FALSE, nbest=2)
  ## print for pedagogical purposes only.  The plot of cp ~ p is more useful.
  ## The line with rss=1e35 is a stepwise() bug, that we reported to S-Plus.
  print(usair.step, digits=4)
  usair.cp <- cp.calc(usair.step, usair, "lnSO2")
  ## print for pedagogical purposes only.  The plot of cp ~ p is more useful.
  usair.cp
  tmp <- (usair.cp$cp <= 10)
  usair.cp[tmp,]

  old.par <- par(mar=par()$mar+c(0,1,0,0))
  tmp <- (usair.cp$cp <= 10)
  plot(cp ~ p, data=usair.cp[tmp,], ylim=c(0,10), type="n", cex=1.3)
  abline(b=1)
  text(x=usair.cp$p[tmp], y=usair.cp$cp[tmp],
       row.names(usair.cp)[tmp], cex=1.3)
  title(main="Cp plot for usair.dat, Cp<10")
  par(old.par)
## export.eps(hh("regb/figure/regb.f3.usair.eps"))
},r={
  usair.regsubset <- leaps::regsubsets(lnSO2~lnmfg+lnpopn+precip+raindays+temp+wind,
                                       data=usair, nbest=2)
  usair.subsets.Summary <- summaryHH(usair.regsubset)
  tmp <- (usair.subsets.Summary$cp <= 10)
  usair.subsets.Summary[tmp,]
  plot(usair.subsets.Summary[tmp,], statistic='cp', legend=FALSE)

  usair.lm7 <- lm.regsubsets(usair.regsubset, 7)
  anova(usair.lm7)
  summary(usair.lm7)
})

vif(lnSO2 ~ temp + lnmfg + lnpopn + wind + precip + raindays, data=usair)

vif(lnSO2 ~ temp + lnmfg + wind + precip, data=usair)

usair.lm <- lm(lnSO2 ~ temp + lnmfg + wind + precip, data=usair)
anova(usair.lm)
summary(usair.lm, corr=FALSE)

Description

Generate a sequence of length points spanning the current.panel.limits()$xlim of a lattice window.

Usage

cplx(length)

Arguments

Value

One-column matrix containing length rows. The first value is the x-value at the left side of the window. The last value is the x-value at the right side of the window. The in between points are evenly spaced.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

scale

Examples

cplx(11)

Description

Most of the datasets are described in the book Statistical Analysis and Data Display.

For ProfChal, see plot.likert.

AudiencePercent is from personal communication by the market researcher who did the study.

SFF8121 is student evaluations of my class compared to the average of all graduate classes in the Spring 2010 semester. Personal communication from the Temple University Office of the Provost to me.

ProfDiv is "Profit-and-Dividend Status of 348 Corportations in the United States for the period from 1929 to 1935" from Brinton WC (1939), Graphic Presentation. Brinton Associates. http://www.archive.org/details/graphicpresentat00brinrich.

NZScienceTeaching is from New Zealand Ministry of Research Science and Technology(2006), “Staying in Science." This URL is no longer valid. http://www.morst.govt.nz/Documents/publications/researchreports/Staying-in-Science-summary.pdf.

PoorChildren is from “Poor Children, Working Parents", Analysis of data from the CensusBureau's American Community Survey. Comparison of Census areas of 100,000 or more people, based on samples from 2005 to 2009.

Source: Data from the U.S. Census Bureau's American Community Survey; analysis by Andrew A. Beveridge, QueensCollege. Copyright 2011 The New York Times Company

https://archive.nytimes.com/www.nytimes.com/imagepages/2011/12/03/opinion/03blow-ch.html?ref=opinion

https://www.nytimes.com/2011/12/03/opinion/blow-newts-war-on-poor-children.html?_r=1

Naomi Robbins and I discuss the PoorChildren example in the Forbes online column: https://www.forbes.com/sites/naomirobbins/2011/12/20/alternative-to-charles-blows-figure-in-newts-war-on-poor-children-2/ demo(PoorChildren, package="HH")

Author(s)

Richard M. Heiberger <[email protected]>

References

Heiberger, Richard M. and Holland, Burt (2015). Statistical Analysis and Data Display: An Intermediate Course with Examples in R. Second Edition. Springer-Verlag, New York. https://link.springer.com/book/10.1007/978-1-4939-2122-5

Description

Intermediate f and chisq functions to simplify writing for both R and S-Plus.

Usage

dchisq.intermediate(x, df, ncp=0, log=FALSE)
pchisq.intermediate(q, df, ncp=0, lower.tail=TRUE, log.p=FALSE)
qchisq.intermediate(p, df, ncp=0, lower.tail=TRUE, log.p=FALSE)
df.intermediate(x, df1, df2, ncp=0, log=FALSE)
pf.intermediate(q, df1, df2, ncp=0, lower.tail=TRUE, log.p=FALSE)
qf.intermediate(p, df1, df2, ncp=0, lower.tail=TRUE, log.p=FALSE)

Arguments

Author(s)

Richard M. Heiberger <[email protected]>

Description

Returns the argument for the diagonal of NA and NULL arguments. For all other arguments, it calls the regular diag function.

Usage

diag.maybe.null(x, ...)

Arguments

Author(s)

Richard M. Heiberger ([email protected])

See Also

diag.

Examples

diag.maybe.null(NULL)
diag.maybe.null(NA)
diag.maybe.null(1:5)

Description

Transpose of ECDF (Empirical CDF) for centered fitted values and residuals from a linear model.

Usage

diagplot5new(linearmodel, ..., pch = 19)

Arguments

Details

This is an implementation in xyplot of the "r-f spread" plot.

Value

"trellis" object.

Author(s)

Richard M. Heiberger <[email protected]>

References

William Cleveland (1993), Visualizing Data, Hobart Press.

Examples

## See ?residVSfitted
## Not run: 
data(fat)
fat.lm <- lm(bodyfat ~ abdomin, data=fat)
diagplot5new(fat.lm)

## End(Not run)

Description

QQ plot of regression residuals. The panel.qqmathline is displayed.

Usage

diagQQ(lm.object, ...)

Arguments

Value

"trellis" object.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

qqmath

Examples

## See ?residVSfitted
## Not run: 
data(fat)
fat.lm <- lm(bodyfat ~ abdomin, data=fat)
diagQQ(fat.lm)

## End(Not run)

Description

Discrete with four levels color dataset. These colors look like four distinct colors when run through the vischeck simulator to see how they look for the three most common color vision deficiencies: Deuteranope, Protanope, Tritanope.

Usage

data("Discrete4")

Format

The format is: chr [1:4] "#E31A1C" "#1F78B4" "#FB9A99" "#A6CEE3"

Details

4x1 color scheme

Examples

data(Discrete4)
## Not run: 
library(RColorBrewer)
library(lattice)
Discrete4 <- brewer.pal(n=12, "Paired")[c(6,2,5,1)]
Discrete4
## save(Discrete4, file="Discrete4.rda") ## data(Discrete4, package="HH")
##
barchart(~ 1:4, col=Discrete4, lwd=0, origin=0, horizontal=FALSE,
         xlab="Colors", scales=list(x=list(labels=Discrete4), y=list(labels=NULL)),
         main=paste("These colors look like four distinct colors when run through",
                    "the vischeck simulator to see how they look for the three most",
                    "common color vision deficiencies: Deuteranope, Protanope, Tritanope.",
                    sep="\n"))

## End(Not run)

Description

Interprets a model formula in the context of its data.frame.

Usage

do.formula.trellis.xysplom(formula, data, na.action = na.pass)

Arguments

Value

A list containing three data.frames and three formula, one for each.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

formula, na.action

Examples

tmp <- data.frame(y=1, x=2, z=3, g=4)
do.formula.trellis.xysplom( y ~ x + z | g, data=tmp)

Description

Helper function for likertWeighted() used for vertical spacing and horizontal borders of grouped panels. Horizontal rules between panels are suppressed by default by likertWeighted unless y.between is non-zero. See examples.

Usage

EmphasizeVerticalPanels(x, y.between)

Arguments

Value

Revised trellis object.

Author(s)

Richard M. Heiberger <[email protected]>

See Also

likertWeighted

Examples

tmp1 <- array(1:60, c(5, 4, 3), list(letters[1:5],letters[6:9],letters[10:12]))
tmp2 <- toCQxR(tmp1)
colnames(tmp2)

likertWeighted(~ . | group + row, tmp2)

likertWeighted(~ . | group + row, tmp2, h.resize.panels=1:5,
               between=list(y=c(0,0,3,0)),
               h.resizePanels=1:5,
               ylab=c("Bottom","Top"),
               xlab.top=c("First","Second","Third"),
               auto.key.title="Response Level",
               main="Three Questions by Five Levels of Classification")

likertObject <- likertWeighted(~ . | group + row, tmp2)
likertObject

EmphasizeVerticalPanels(likertObject, y.between=c(0,0,1,0))

Description

Remove main title, left axis tick labels, left strip, bottom legend from plot and keep the vertical spacing allocated to those items. This function is used to prepare a trellis object to be placed next to another trellis object. The current object will have much of its annotation removed with the intent of sharing annotation with the other object. This is motivated by the ProfChal example in plot.likert.

Usage

emptyMainLeftAxisLeftStripBottomLegend(x)

Arguments

Details

We manipulate the items inside the trellis object.

Value

A "trellis" object with the stated items replaced by non-printing values. The vertical spacing of the original object is retained.

Author(s)

Richard M. Heiberger <[email protected]>

References

The manipulations are similar to those in the c.trellis and related functions in the latticeExtra package.

See Also

plot.likert

Examples

## This is a small example.
## See ?plot.likert for the complete example including motivation.
##
require(grid)
require(lattice)
require(latticeExtra)
require(HH)

data(ProfChal)

tmp <- data.matrix(ProfChal[,1:5])
rownames(tmp) <- ProfChal$Question
ProfChal.list <- split.data.frame(tmp, ProfChal$Subtable)

Empl <- ProfChal.list[[2]]

pct   <- likert(Empl, as.percent="noRightAxis", xlab="Percent")
pct
count <- likert(Empl, rightAxis=TRUE,
                xlab="Count", ylab.right="Row Count Totals",
                scales=list(x=list(at=c(0, 100, 200))))
count
countEmptied <- HH:::emptyMainLeftAxisLeftStripBottomLegend(count)
countEmptied

tmp <- update(resizePanels(c(pct, countEmptied, y.same=TRUE, layout=c(2,1)), w=c(.8, .2)),
              scales=list(y=list(alternating=3, limits=count$y.limits),
                          x=list(at=list(pct$x.scales$at, count$x.scales$at),
                                 labels=list(pct$x.scales$labels,
                                             count$x.scales$labels))),
              xlab=c(" ", pct$xlab, " ", count$xlab),
              between=list(x=1))
tmp

Description

Exports a graph from the current device in R, or the graphsheet in S-Plus, to an EPS file.

Usage

export.eps(FileName.in, Name.in="GSD2", ...)

Arguments

Author(s)

Richard M. Heiberger <[email protected]>

See Also

dev2.

Examples

## Not run: 
  if (interactive()) {
    trellis.device()
    plot(1:10)
    export.eps("abcd.eps")
  }
  
## End(Not run)

Description

Miscellaneous functions that I wish were in or consistent between S-Plus and R.

Usage

as.rts(x, ...)

title.trellis(main = NULL, sub = NULL, xlab = NULL, ylab = NULL,
    line = NA, outer = FALSE, axes=NULL, ...)

title.grob(main=NULL, y=.99, gp=gpar(cex=1.5))

## S3 method for class 'arima.model'
as.character(x, ...)

arima.model(x)

coefArimaHH(object, ...)

.arima.info.names.not.ordered (model)

Arguments

Value

The result object of arima.model has class "arima.model"

Author(s)

Richard M. Heiberger ([email protected])

See Also

arma.loop

Description

Plot a chisquare or a F-curve. Shade a region for rejection region or do-not-reject region. F.observed and chisq.observed plots a vertical line with arrowhead markers at the location of the observed xbar and outlines the area corresponding to the $p$-value.

Usage

F.setup(df1=1,
        df2=Inf,
        ncp=0,
        log.p=FALSE,
        xlim.in=c(0, 5),
        ylim.in=range(c(0, 1.1*df.intermediate(x=seq(.5,1.5,.01),
                       df1=df1, df2=df2, ncp=ncp, log=log.p))),
        main.in=main.calc, ylab.in="F density",
        ...)

F.curve(df1=1,
        df2=Inf,
        ncp=0,
        log.p=FALSE,
        alpha=.05,
        critical.values=f.alpha,
        f=seq(0, par()$usr[2], length=109),
        shade="right", col=par("col"),
        axis.name="f",
        ...)

F.observed(f.obs, col="green",
           df1=1,
           df2=Inf,
           ncp=0,
           log.p=FALSE,
           axis.name="f",
           shade="right",
           shaded.area=0,
           display.obs=TRUE)

chisq.setup(df=1,
           ncp=0,
           log.p=FALSE,
           xlim.in=c(0, qchisq.intermediate(p=1-.01, df=df, ncp=ncp, log.p=log.p)),
           ylim.in=range(c(0, 1.1*dchisq.intermediate(x=seq(max(0.5,df-2),df+2,.01),
                         df=df, ncp=ncp, log=log.p))),
           main.in=main.calc, ylab.in="Chisq density",
           ...)

chisq.curve(df=1,
            ncp=0,
            log.p=FALSE,
            alpha=.05,
            critical.values=chisq.alpha,
            chisq=seq(0, par()$usr[2], length=109),
            shade="right", col=par("col"),
            axis.name="chisq",
            ...)

chisq.observed(chisq.obs, col="green",
               df=1,
               ncp=0,
               log.p=FALSE,
               axis.name="chisq",
               shade="right",
               shaded.area=0,
               display.obs=TRUE)

Arguments

Author(s)

Richard M. Heiberger <[email protected]>

Examples

old.omd <- par(omd=c(.05,.88, .05,1))
chisq.setup(df=12)
chisq.curve(df=12, col='blue')
chisq.observed(22, df=12)
par(old.omd)

old.omd <- par(omd=c(.05,.88, .05,1))
chisq.setup(df=12)
chisq.curve(df=12, col='blue', alpha=c(.05, .05))
par(old.omd)

old.omd <- par(omd=c(.05,.88, .05,1))
F.setup(df1=5, df2=30)
F.curve(df1=5, df2=30, col='blue')
F.observed(3, df1=5, df2=30)
par(old.omd)

old.omd <- par(omd=c(.05,.88, .05,1))
F.setup(df1=5, df2=30)
F.curve(df1=5, df2=30, col='blue', alpha=c(.05, .05))
par(old.omd)