Package 'vcd'

Description

Representation of a $k \times k$ confusion matrix, where the observed and expected diagonal elements are represented by superposed black and white rectangles, respectively. The function also computes a statistic measuring the strength of agreement (relation of respective area sums).

Usage

## Default S3 method:
agreementplot(x, reverse_y = TRUE, main = NULL,
              weights = c(1, 1 - 1/(ncol(x) - 1)^2), margins = par("mar"),
              newpage = TRUE, pop = TRUE, 
              xlab = names(dimnames(x))[2],
              ylab = names(dimnames(x))[1],
              xlab_rot = 0, xlab_just = "center",
              ylab_rot = 90, ylab_just = "center",
              fill_col = function(j) gray((1 - (weights[j]) ^ 2) ^ 0.5),
              line_col = "red", xscale = TRUE, yscale = TRUE,
              return_grob = FALSE,
              prefix = "", ...)
## S3 method for class 'formula'
agreementplot(formula, data = NULL, ..., subset)

Arguments

Details

Weights can be specified to allow for partial agreement, taking into account contributions from off-diagonal cells. Partial agreement is typically represented in the display by lighter shading, as given by fill_col(j), corresponding to weights[j].

A weight vector of length 1 means strict agreement only, each additional element increases the maximum number of disagreement steps.

cotabplot can be used for stratified analyses (see examples).

Value

Invisibly returned, a list with components

Author(s)

David Meyer [email protected]

References

Bangdiwala, S. I. (1988). The Agreement Chart. Department of Biostatistics, University of North Carolina at Chapel Hill, Institute of Statistics Mimeo Series No. 1859, https://repository.lib.ncsu.edu/bitstreams/fea554e9-8750-4f1a-8419-ee126ce1a790/download

Bangdiwala, S. I., Ana S. Haedo, Marcela L. Natal, and Andres Villaveces. The agreement chart as an alternative to the receiver-operating characteristic curve for diagnostic tests. Journal of Clinical Epidemiology, 61 (9), 866-874.

Michael Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

data("SexualFun")
agreementplot(t(SexualFun))

data("MSPatients")
## Not run: 
## best visualized using a resized device, e.g. using:
## get(getOption("device"))(width = 12)
pushViewport(viewport(layout = grid.layout(ncol = 2)))
pushViewport(viewport(layout.pos.col = 1))
agreementplot(t(MSPatients[,,1]), main = "Winnipeg Patients",
              newpage = FALSE)
popViewport()
pushViewport(viewport(layout.pos.col = 2))
agreementplot(t(MSPatients[,,2]), main = "New Orleans Patients",
              newpage = FALSE)
popViewport(2)
dev.off()

## End(Not run)

## alternatively, use cotabplot:
cotabplot(MSPatients, panel = cotab_agreementplot)

Description

Data from Koch & Edwards (1988) from a double-blind clinical trial investigating a new treatment for rheumatoid arthritis.

Usage

data("Arthritis")

Format

A data frame with 84 observations and 5 variables.

ID

patient ID.

Treatment

factor indicating treatment (Placebo, Treated).

Sex

factor indicating sex (Female, Male).

Age

age of patient.

Improved

ordered factor indicating treatment outcome (None, Some, Marked).

Source

Michael Friendly (2000), Visualizing Categorical Data: http://euclid.psych.yorku.ca/ftp/sas/vcd/catdata/arthrit.sas

References

G. Koch & S. Edwards (1988), Clinical efficiency trials with categorical data. In K. E. Peace (ed.), Biopharmaceutical Statistics for Drug Development, 403–451. Marcel Dekker, New York.

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

data("Arthritis")
art <- xtabs(~ Treatment + Improved, data = Arthritis, subset = Sex == "Female")
art

mosaic(art, gp = shading_Friendly)
mosaic(art, gp = shading_max)

Description

Produce an association plot indicating deviations from a specified independence model in a possibly high-dimensional contingency table.

Usage

## Default S3 method:
assoc(x, row_vars = NULL, col_vars = NULL, compress = TRUE,
  xlim = NULL, ylim = NULL,
  spacing = spacing_conditional(sp = 0), spacing_args = list(),
  split_vertical = NULL, keep_aspect_ratio = FALSE, 
  xscale = 0.9, yspace = unit(0.5, "lines"), main = NULL, sub = NULL,
  ..., residuals_type = "Pearson", gp_axis = gpar(lty = 3))
## S3 method for class 'formula'
assoc(formula, data = NULL, ..., subset = NULL, na.action = NULL, main = NULL, sub = NULL)

Arguments

Details

Association plots have been suggested by Cohen (1980) and extended by Friendly (1992) and provide a means for visualizing the residuals of an independence model for a contingency table.

assoc is a generic function and currently has a default method and a formula interface. Both are high-level interfaces to the strucplot function, and produce (extended) association plots. Most of the functionality is described there, such as specification of the independence model, labeling, legend, spacing, shading, and other graphical parameters.

For a contingency table, the signed contribution to Pearson's $\chi^2$ for cell $\{ij\ldots k\}$ is

$d_{ij\ldots k} = \frac{(f_{ij\ldots k} - e_{ij\ldots k})}{ \sqrt{e_{ij\ldots k}}}$

where $f_{ij\ldots k}$ and $e_{ij\ldots k}$ are the observed and expected counts corresponding to the cell. In the association plot, each cell is represented by a rectangle that has (signed) height proportional to $d_{ij\ldots k}$ and width proportional to $\sqrt{e_{ij\ldots k}}$, so that the area of the box is proportional to the difference in observed and expected frequencies. The rectangles in each row are positioned relative to a baseline indicating independence ($d_{ij\ldots k} = 0$). If the observed frequency of a cell is greater than the expected one, the box rises above the baseline, and falls below otherwise.

Additionally, the residuals can be colored depending on a specified shading scheme (see Meyer et al., 2003). Package vcd offers a range of residual-based shadings (see the shadings help page). Some of them allow, e.g., the visualization of test statistics.

Unlike the assocplot function in the graphics package, this function allows the visualization of contingency tables with more than two dimensions. Similar to the construction of ‘flat’ tables (like objects of class "ftable" or "structable"), the dimensions are folded into rows and columns.

The layout is very flexible: the specification of shading, labeling, spacing, and legend is modularized (see strucplot for details).

Value

The "structable" visualized is returned invisibly.

Author(s)

David Meyer [email protected]

References

Cohen, A. (1980), On the graphical display of the significant components in a two-way contingency table. Communications in Statistics—Theory and Methods, A9, 1025–1041.

Friendly, M. (1992), Graphical methods for categorical data. SAS User Group International Conference Proceedings, 17, 190–200. http://datavis.ca/papers/sugi/sugi17.pdf

Meyer, D., Zeileis, A., Hornik, K. (2003), Visualizing independence using extended association plots. Proceedings of the 3rd International Workshop on Distributed Statistical Computing, K. Hornik, F. Leisch, A. Zeileis (eds.), ISSN 1609-395X. https://www.R-project.org/conferences/DSC-2003/Proceedings/

Meyer, D., Zeileis, A., and Hornik, K. (2006), The strucplot framework: Visualizing multi-way contingency tables with vcd. Journal of Statistical Software, 17(3), 1-48. doi:10.18637/jss.v017.i03 and available as vignette ("strucplot", package = "vcd").

See Also

mosaic, strucplot, structable

Examples

data("HairEyeColor")
## Aggregate over sex:
(x <- margin.table(HairEyeColor, c(1, 2)))

## Ordinary assocplot:
assoc(x)
## and with residual-based shading (of independence)
assoc(x, main = "Relation between hair and eye color", shade = TRUE)

## Aggregate over Eye color:
(x <- margin.table(HairEyeColor, c(1, 3)))
chisq.test(x)
assoc(x, main = "Relation between hair color and sex", shade = TRUE)

# Visualize multi-way table
assoc(aperm(HairEyeColor), expected = ~ (Hair + Eye) * Sex,
      labeling_args = list(just_labels = c(Eye = "left"),
                           offset_labels = c(right = -0.5),
                           offset_varnames = c(right = 1.2),
                           rot_labels = c(right = 0),
                           tl_varnames = c(Eye = TRUE))
)

assoc(aperm(UCBAdmissions), expected = ~ (Admit + Gender) * Dept, compress = FALSE,
      labeling_args = list(abbreviate_labs = c(Gender = TRUE), rot_labels = 0)
)

Description

Computes the Pearson chi-Squared test, the Likelihood Ratio chi-Squared test, the phi coefficient, the contingency coefficient and Cramer's V for possibly stratified contingency tables.

Usage

assocstats(x)

Arguments

Value

In case of a 2-dimensional table, a list with components:

In case of higher-dimensional tables, a list of the above mentioned structure, each list component representing one stratum defined by the combinations of all levels of the stratum dimensions.

Author(s)

David Meyer [email protected]

References

Michael Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Fleiss, J. L. (1981). Statistical methods for rates and proportions (2nd ed). New York: Wiley

Examples

data("Arthritis")
tab <- xtabs(~Improved + Treatment, data = Arthritis)
summary(assocstats(tab))

assocstats(UCBAdmissions)

Description

Baseball data.

Usage

data("Baseball")

Format

A data frame with 322 observations and 25 variables.

name1

player's first name.

name2

player's last name.

atbat86

times at Bat: number of official plate appearances by a hitter. It counts as an official at-bat as long as the batter does not walk, sacrifice, get hit by a pitch or reach base due to catcher's interference.

hits86

hits.

homer86

home runs.

runs86

the number of runs scored by a player. A run is scored by an offensive player who advances from batter to runner and touches first, second, third and home base in that order without being put out.

rbi86

Runs Batted In: A hitter earns a run batted in when he drives in a run via a hit, walk, sacrifice (bunt or fly) fielder's choice, hit-batsman or on an error (when the official scorer rules that the run would have scored anyway).

walks86

A “walk” (or “base on balls”) is an award of first base granted to a batter who receives four pitches outside the strike zone.

years

Years in the Major Leagues. Seems to count all years a player has actually played in the Major Leagues, not necessarily consecutive.

atbat

career times at bat.

hits

career hits.

homeruns

career home runs.

runs

career runs.

rbi

career runs batted in.

walks

career walks.

league86

player's league.

div86

player's division.

team86

player's team.

posit86

player's position (see Hitters).

outs86

number of putouts (see Hitters)

assist86

number of assists (see Hitters)

error86

number of assists (see Hitters)

sal87

annual salary on opening day (in USD 1000).

league87

league in 1987.

team87

team in 1987.

Source

SAS System for Statistical Graphics, First Edition, page A2.3

References

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

See Also

Hitters

Examples

data("Baseball")

Description

Creates a display of observed and fitted values for a binary regression model with one numeric predictor, conditioned by zero or many co-factors.

Usage

binreg_plot(model, main = NULL, xlab = NULL, ylab = NULL,
            xlim = NULL, ylim = NULL,
            pred_var = NULL, pred_range = c("data", "xlim"),
            group_vars = NULL, base_level = NULL, subset,
            type = c("response", "link"), conf_level = 0.95, delta = FALSE,
            pch = NULL, cex = 0.6, jitter_factor = 0.1,
            lwd = 5, lty = 1, point_size = 0, col_lines = NULL, col_bands = NULL, 
            legend = TRUE, legend_pos = NULL, legend_inset = c(0, 0.1),
            legend_vgap = unit(0.5, "lines"),
            labels = FALSE, labels_pos = c("right", "left"),
            labels_just = c("left","center"), labels_offset = c(0.01, 0),
            gp_main = gpar(fontface = "bold", fontsize = 14),
            gp_legend_frame = gpar(lwd = 1, col = "black"),
            gp_legend_title = gpar(fontface = "bold"),
            newpage = TRUE, pop = FALSE, return_grob = FALSE)
grid_abline(a, b, ...)

Arguments

Details

The primary purpose of binreg_plot() is to visualize observed and fitted values for binary regression models (like the logistic or probit regression model) with one numeric predictor. If one or more categorical predictors are used in the model, the fitted values are conditioned on them, i.e. separate curves are drawn corresponding to the factor level combinations. Thus, it shows a full-model plot, not a conditional plot where several models would be fit to data subsets.

The implementation relies on objects returned by glm, as it uses its "terms" and "model" components.

The function tries to determine suitable values for the legend and/or labels, but depending on the data, this might require some tweaking.

By default, the limits of the confidence band are determined for the linear predictor (i.e., on the link scale) and transformed to response scale (if this is the chosen plot type) using the inverse link function. If delta is TRUE, the limits are determined on the response scale. Note that the resulting band using the delta method is symmetric around the fitted mean, but may exceed the unit interval (on the response scale) and will be cut off.

grid_abline() is a simple convenience wrapper for grid.abline with similar behavior than abline in that it extracts coefficients from a regression model, if given instead of the intercept a.

Value

if return_grob is TRUE, a grob object corresponding to the plot. NULL (invisibly) else.

Author(s)

David Meyer [email protected]

References

Michael Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

## Simple model with no conditioning variables
art.mod0 <- glm(Improved > "None" ~ Age, data = Arthritis, family = binomial)

binreg_plot(art.mod0, "Arthritis Data")
binreg_plot(art.mod0, type = "link") ## logit scale

## one conditioning factor
art.mod1 <- update(art.mod0, . ~ . + Sex)
binreg_plot(art.mod1)
binreg_plot(art.mod1, legend = FALSE, labels = TRUE, xlim = c(20, 80))

## two conditioning factors
art.mod2 <- update(art.mod1, . ~ . + Treatment)
binreg_plot(art.mod2)
binreg_plot(art.mod2, subset = Sex == "Male") ## subsetting

## some tweaking
binreg_plot(art.mod2, gp_legend_frame = gpar(col = NA, fill = "white"), col_bands = NA)
binreg_plot(art.mod2, legend = FALSE, labels = TRUE,
            labels_pos = "left", labels_just = c("left", "top"))

## model with grouped response data
shuttle.mod <- glm(cbind(nFailures, 6 - nFailures) ~ Temperature,
                   data = SpaceShuttle, na.action = na.exclude, family = binomial)
binreg_plot(shuttle.mod, xlim = c(30, 81), pred_range = "xlim",
            ylab = "O-Ring Failure Probability", xlab = "Temperature (F)")

Description

Data from the Danish Welfare Study about broken marriages or permanent relationships depending on gender and social rank.

Usage

data("BrokenMarriage")

Format

A data frame with 20 observations and 4 variables.

Freq

frequency.

gender

factor indicating gender (male, female).

rank

factor indicating social rank (I, II, III, IV, V).

broken

factor indicating whether the marriage or permanent relationship was broken (yes, no).

Source

E. B. Andersen (1991), The Statistical Analysis of Categorical Data, page 177.

References

E. B. Andersen (1991), The Statistical Analysis of Categorical Data. 2nd edition. Springer-Verlag, Berlin.

Examples

data("BrokenMarriage")
structable(~ ., data = BrokenMarriage)

Description

Results from the first German soccer league (1963-2008).

Usage

data("Bundesliga")

Format

A data frame with 14018 observations and 7 variables.

HomeTeam

factor. Name of the home team.

AwayTeam

factor. Name of the away team.

HomeGoals

number of goals scored by the home team.

AwayGoals

number of goals scored by the away team.

Round

round of the game.

Year

year in which the season started.

Date

starting time of the game (in "POSIXct" format).

Details

The data comprises all games in the first German soccer league since its foundation in 1963. The data have been queried online from the official Web page of the DFB and prepared as a data frame in R by Daniel Dekic, Torsten Hothorn, and Achim Zeileis (replacing earlier versions of the data in the package containing only subsets of years).

Each year/season comprises 34 rounds (except 1963, 1964, 1991) so that all 18 teams play twice against each other (switching home court advantage). In 1963/64, there were only 16 teams, hence only 30 rounds. In 1991, after the German unification, there was one season with 20 teams and 38 rounds.

Source

Homepage of the Deutscher Fussball-Bund (DFB, German Football Association): https://www.dfb.de/index/

References

Leonhard Knorr-Held (1999), Dynamic rating of sports teams. SFB 386 “Statistical Analysis of Discrete Structures”, Discussion paper 98.

See Also

UKSoccer

Examples

data("Bundesliga")

## number of goals per game poisson distributed?
ngoals1 <- xtabs(~ HomeGoals, data = Bundesliga, subset = Year == 1995)
ngoals2 <- xtabs(~ AwayGoals, data = Bundesliga, subset = Year == 1995)
ngoals3 <- table(apply(subset(Bundesliga, Year == 1995)[,3:4], 1, sum))

gf1 <- goodfit(ngoals1)
gf2 <- goodfit(ngoals2)
gf3 <- goodfit(ngoals3)

summary(gf1)
summary(gf2)
summary(gf3)
plot(gf1)
plot(gf2)
plot(gf3)

Ord_plot(ngoals1)
distplot(ngoals1)

Description

Number of votes by province in the German Bundestag election 2005 (for the parties that eventually entered the parliament).

Usage

data("Bundestag2005")

Format

A 2-way "table" giving the number of votes for each party (Fraktion) in each of the 16 German provinces (Bundesland):

Details

In the election for the German parliament “Bundestag”, five parties obtained enough votes to enter the parliament: the social democrats SPD, the conservative CDU/CSU, the liberal FDP, the green party “Die Gruenen” and the leftist party “Die Linke”. The table Bundestag2005 gives the number of votes for each party (Fraktion) in each of the 16 German provinces (Bundesland). The provinces are ordered from North to South.

The data have been obtained from the German statistical office (Statistisches Bundesamt) from the Web page given below.

Note that the number of seats in the parliament cannot be computed from the number of votes alone. The examples below show the distribution of seats that resulted from the election.

Source

Die Bundeswahlleiterin, Statistisches Bundesamt. https://www.bundeswahlleiterin.de/bundestagswahlen/2005.html

Examples

library(colorspace)
## The outcome of the election in terms of seats in the
## parliament was:
seats <- structure(c(226, 61, 54, 51, 222),
  .Names = c("CDU/CSU", "FDP",  "Linke", "Gruene", "SPD"))

## Hues are chosen as metaphors for the political parties
## CDU/CSU: blue, FDP: yellow, Linke: purple, Gruene: green, SPD: red
## using the respective hues from a color wheel with
## chroma = 60 and luminance = 75
parties <- rainbow_hcl(6, c = 60, l = 75)[c(5, 2, 6, 3, 1)]
names(parties) <- names(seats)
parties

## The pie chart shows that neither the SPD+Gruene coalition nor
## the opposition of CDU/CSU+FDP could assemble a majority.
## No party would enter a coalition with the leftists, leading to a
## big coalition.
pie(seats, clockwise = TRUE, col = parties)

## The regional distribution of the votes, stratified by province,
## is shown in a mosaic display: first for the 10 Western then the
## 6 Eastern provinces.
data("Bundestag2005")
votes <- Bundestag2005[c(1, 3:5, 9, 11, 13:16, 2, 6:8, 10, 12),
                       c("CDU/CSU", "FDP", "SPD", "Gruene", "Linke")]
mosaic(votes, gp = gpar(fill = parties[colnames(votes)]),
  spacing = spacing_highlighting, labeling = labeling_left,
  labeling_args = list(rot_labels = c(0, 90, 0, 0), pos_labels = "center",
  just_labels = c("center","center","center","right"), varnames = FALSE),
  margins = unit(c(2.5, 1, 1, 12), "lines"),
  keep_aspect_ratio = FALSE)

Description

Data from Fisher et al. (1943) giving the number of tokens found for each of 501 species of butterflies collected in Malaya.

Usage

data("Butterfly")

Format

A 1-way table giving the number of tokens for 501 species of butterflies. The variable and its levels are

Source

Michael Friendly (2000), Visualizing Categorical Data, pages 21–22.

References

R. A. Fisher, A. S. Corbet, C. B. Williams (1943), The relation between the number of species and the number of individuals, Journal of Animal Ecology, 12, 42–58.

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

data("Butterfly")
Ord_plot(Butterfly)

Description

Computes and plots conditional densities describing how the distribution of a categorical variable y changes over a numerical variable x.

Usage

cd_plot(x, ...)
## Default S3 method:
cd_plot(x, y,
  plot = TRUE, ylab_tol = 0.05,
  bw = "nrd0", n = 512, from = NULL, to = NULL,
  main = "", xlab = NULL, ylab = NULL, margins = c(5.1, 4.1, 4.1, 3.1),
  gp = gpar(), name = "cd_plot", newpage = TRUE, pop = TRUE, return_grob = FALSE, ...)
## S3 method for class 'formula'
cd_plot(formula, data = list(),
  plot = TRUE, ylab_tol = 0.05,
  bw = "nrd0", n = 512, from = NULL, to = NULL,
  main = "", xlab = NULL, ylab = NULL, margins = c(5.1, 4.1, 4.1, 3.1),
  gp = gpar(), name = "cd_plot", newpage = TRUE, pop = TRUE, return_grob = FALSE, ...)

Arguments

Details

cd_plot computes the conditional densities of x given the levels of y weighted by the marginal distribution of y. The densities are derived cumulatively over the levels of y.

This visualization technique is similar to spinograms (see spine) but they do not discretize the explanatory variable, but rather use a smoothing approach. Furthermore, the original x axis and not a distorted x axis (as for spinograms) is used. This typically results in conditional densities that are based on very few observations in the margins: hence, the estimates are less reliable there.

Value

The conditional density functions (cumulative over the levels of y) are returned invisibly.

Author(s)

Achim Zeileis [email protected]

References

Hofmann, H., Theus, M. (2005), Interactive graphics for visualizing conditional distributions, Unpublished Manuscript.

See Also

spine, density

Examples

## Arthritis data
data("Arthritis")
cd_plot(Improved ~ Age, data = Arthritis)
cd_plot(Improved ~ Age, data = Arthritis, bw = 3)
cd_plot(Improved ~ Age, data = Arthritis, bw = "SJ")
## compare with spinogram
spine(Improved ~ Age, data = Arthritis, breaks = 3)

## Space shuttle data
data("SpaceShuttle")
cd_plot(Fail ~ Temperature, data = SpaceShuttle, bw = 2)

## scatter plot with conditional density
cdens <- cd_plot(Fail ~ Temperature, data = SpaceShuttle, bw = 2, plot = FALSE)
plot(I(-1 * (as.numeric(Fail) - 2)) ~ jitter(Temperature, factor = 2), data = SpaceShuttle,
  xlab = "Temperature", ylab = "Failure")
lines(53:81, cdens[[1]](53:81), col = 2)

Description

For a contingency table in array form, compute a list of conditional tables given some margins.

Usage

co_table(x, margin, collapse = ".")

Arguments

Details

This is essentially an interface to [ which is more convenient for arrays of arbitrary dimension.

Value

A list of the resulting conditional tables.

Author(s)

Achim Zeileis [email protected]

Examples

data("HairEyeColor")
co_table(HairEyeColor, 1)
co_table(HairEyeColor, c("Hair", "Eye"))
co_table(HairEyeColor, 1:2, collapse = "")

Description

Data from Ashford & Sowden (1970) given by Agresti (1990) on the association between two pulmonary conditions, breathlessness and wheeze, in a large sample of coal miners who were smokers with no radiological evidence of pneumoconlosis, aged between 20–64 when examined. This data is frequently used as an example of fitting models for bivariate, binary responses.

Usage

data("CoalMiners")

Format

A 3-dimensional table of size 2 x 2 x 9 resulting from cross-tabulating variables for 18,282 coal miners. The variables and their levels are as follows:

Details

In an earlier version of this data set, the first group, aged 20-24, was inadvertently omitted from this data table and the breathlessness variable was called wheeze and vice versa.

Source

Michael Friendly (2000), Visualizing Categorical Data, pages 82–83, 319–322.

References

A. Agresti (1990), Categorical Data Analysis. Wiley-Interscience, New York, Table 7.11, p. 237

J. R. Ashford and R. D. Sowdon (1970), Multivariate probit analysis, Biometrics, 26, 535–546.

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

data("CoalMiners")

ftable(CoalMiners, row.vars = 3)

## Fourfold display, both margins equated
fourfold(CoalMiners[,,2:9], mfcol = c(2,4))

## Fourfold display, strata equated
fourfold(CoalMiners[,,2:9], std = "ind.max", mfcol = c(2,4))


## Log Odds Ratio Plot
lor_CM <- loddsratio(CoalMiners)
summary(lor_CM)
plot(lor_CM)
lor_CM_df <- as.data.frame(lor_CM)

# fit linear models using WLS
age <- seq(20, 60, by = 5)
lmod <- lm(LOR ~ age, weights = 1 / ASE^2, data = lor_CM_df)
grid.lines(age, fitted(lmod), gp = gpar(col = "blue"))
qmod <- lm(LOR ~ poly(age, 2), weights = 1 / ASE^2, data = lor_CM_df)
grid.lines(age, fitted(qmod), gp = gpar(col = "red"))

Description

Performs a test of (conditional) independence of 2 margins in a contingency table by simulation from the marginal distribution of the input table under (conditional) independence.

Usage

coindep_test(x, margin = NULL, n = 1000, 
  indepfun = function(x) max(abs(x)), aggfun = max,
  alternative = c("greater", "less"),
  pearson = TRUE)

Arguments

Details

If margin is NULL this computes a simple independence statistic in a 2-way table. Alternatively, margin can give several conditioning variables and then conditional independence in the resulting conditional table is tested.

By default, this uses a (double) maximum statistic of Pearson residuals. By changing indepfun or aggfun a (maximum of) Pearson Chi-squared statistic(s) can be computed or just the usual Pearson Chi-squared statistics and so on. Other statistics can be computed by changing pearson to FALSE.

The function uses r2dtable to simulate the distribution of the test statistic under the null.

Value

A list of class "coindep_test" inheriting from "htest" with following components:

Author(s)

Achim Zeileis [email protected]

See Also

chisq.test, fisher.test, r2dtable

Examples

library(MASS)
TeaTasting <- matrix(c(3, 1, 1, 3), nrow = 2,
                     dimnames = list(Guess = c("Milk", "Tea"),
                                     Truth = c("Milk", "Tea"))
)
## compute maximum statistic
coindep_test(TeaTasting)
## compute Chi-squared statistic
coindep_test(TeaTasting, indepfun = function(x) sum(x^2))
## use unconditional asymptotic distribution
chisq.test(TeaTasting, correct = FALSE)
chisq.test(TeaTasting)


data("UCBAdmissions")
## double maximum statistic
coindep_test(UCBAdmissions, margin = "Dept")
## maximum of Chi-squared statistics
coindep_test(UCBAdmissions, margin = "Dept", indepfun = function(x) sum(x^2))
## Pearson Chi-squared statistic
coindep_test(UCBAdmissions, margin = "Dept", indepfun = function(x) sum(x^2), aggfun = sum)
## use unconditional asymptotic distribution
loglm(~ Dept * (Gender + Admit), data = UCBAdmissions)

Description

Panel-generating functions visualizing contingency tables that can be passed to cotabplot.

Usage

cotab_mosaic(x = NULL, condvars = NULL, ...)
cotab_assoc(x = NULL, condvars = NULL, ylim = NULL, ...)
cotab_sieve(x = NULL, condvars = NULL, ...)
cotab_loddsratio(x = NULL, condvars = NULL, ...)
cotab_agreementplot(x = NULL, condvars = NULL, ...)
cotab_fourfold(x = NULL, condvars = NULL, ...)
cotab_coindep(x, condvars,
  test = c("doublemax", "maxchisq", "sumchisq"),
  level = NULL, n = 1000, interpolate = c(2, 4),
  h = NULL, c = NULL, l = NULL, lty = 1,
  type = c("mosaic", "assoc"), legend = FALSE, ylim = NULL, ...)

Arguments

Details

These functions of class "panel_generator" are panel-generating functions for use with cotabplot, i.e., they return functions with the interface

panel(x, condlevels)

required for cotabplot. The functions produced by cotab_mosaic, cotab_assoc and cotab_sieve essentially only call co_table to produce the conditioned table and then call mosaic, assoc or sieve respectively with the arguments specified.

The function cotab_coindep is similar but additionally chooses an appropriate residual-based shading visualizing the associated conditional independence model. The conditional independence test is carried out via coindep_test and the shading is set up via shading_hcl.

A description of the underlying ideas is given in Zeileis, Meyer, Hornik (2005).

Author(s)

Achim Zeileis [email protected]

References

Meyer, D., Zeileis, A., and Hornik, K. (2006), The strucplot framework: Visualizing multi-way contingency tables with vcd. Journal of Statistical Software, 17(3), 1-48. doi:10.18637/jss.v017.i03 and available as vignette("strucplot").

Zeileis, A., Meyer, D., Hornik K. (2007), Residual-based shadings for visualizing (conditional) independence, Journal of Computational and Graphical Statistics, 16, 507–525.

See Also

cotabplot, mosaic, assoc, sieve, co_table, coindep_test, shading_hcl

Examples

data("UCBAdmissions")

cotabplot(~ Admit + Gender | Dept, data = UCBAdmissions)
cotabplot(~ Admit + Gender | Dept, data = UCBAdmissions, panel = cotab_assoc)
cotabplot(~ Admit + Gender | Dept, data = UCBAdmissions, panel = cotab_fourfold)

ucb <- cotab_coindep(UCBAdmissions, condvars = "Dept", type = "assoc",
                     n = 5000, margins = c(3, 1, 1, 3))
cotabplot(~ Admit + Gender | Dept, data = UCBAdmissions, panel = ucb)

Description

cotabplot is a generic function for creating trellis-like coplots (conditional plots) for contingency tables.

Usage

cotabplot(x, ...)
## Default S3 method:
cotabplot(x, cond = NULL,
  panel = cotab_mosaic, panel_args = list(),
  margins = rep(1, 4), layout = NULL,
  text_gp = gpar(fontsize = 12), rect_gp = gpar(fill = grey(0.9)),
  pop = TRUE, newpage = TRUE, return_grob = FALSE,
  ...)
## S3 method for class 'formula'
cotabplot(formula, data = NULL, ...)

Arguments

Details

cotabplot is a generic function designed to create coplots or conditional plots (see Cleveland, 1993, and Becker, Cleveland, Shyu, 1996) similar to coplot but for contingency tables.

cotabplot takes on computing the conditioning information and setting up the trellis display, and then relies on a panel function to create plots from the full table and the conditioning information. A simple example would be a contingency table tab with margin names "x", "y" and "z". To produce this plot either the default interface can be used or the formula interface via

cotabplot(tab, "z") cotabplot(~ x + y | z, data = tab)

The panel function needs to be of the form

panel(x, condlevels)

where x is the full table (tab in the example above) and condlevels is a named vector with the levels (e.g., c(z = "z1") in the example above).

Alternatively, panel can also be a panel-generating function of class "grapcon_generator" which creates a function with the interface described above. The panel-generating function is called with the interface

panel(x, condvars, ...)

where again x is the full table, condvars is now only a vector with the names of the conditioning variables (and not their levels, e.g., "z" in the example above). Further arguments can be passed to the panel-generating function via ... which also includes the arguments set in panel_args.

Suitable panel-generating functions for mosaic, association and sieve plots can be found at cotab_mosaic.

A description of the underlying ideas is given in Zeileis, Meyer, Hornik (2005).

Author(s)

Achim Zeileis [email protected]

References

Becker, R.A., Cleveland, W.S., Shyu, M.-J. (1996), The visual design and control of trellis display. Journal of Computational and Graphical Statistics, 5, 123–155.

Cleveland, W.S. (1993), Visualizing Data, Summit, New Jersey: Hobart Press.

Meyer, D., Zeileis, A., and Hornik, K. (2006), The strucplot framework: Visualizing multi-way contingency tables with vcd. Journal of Statistical Software, 17(3), 1-48. doi:10.18637/jss.v017.i03 and available as vignette("strucplot").

Zeileis, A., Meyer, D., Hornik K. (2007), Residual-based shadings for visualizing (conditional) independence, Journal of Computational and Graphical Statistics, 16, 507–525.

See Also

cotab_mosaic, cotab_coindep, co_table, coindep_test

Examples

data("UCBAdmissions")

cotabplot(~ Admit + Gender | Dept, data = UCBAdmissions)
cotabplot(~ Admit + Gender | Dept, data = UCBAdmissions, panel = cotab_assoc)

ucb <- cotab_coindep(UCBAdmissions, condvars = "Dept", type = "assoc",
                     n = 5000, margins = c(3, 1, 1, 3))
cotabplot(~ Admit + Gender | Dept, data = UCBAdmissions, panel = ucb)

Description

Data from the Danish Welfare Study.

Usage

data("DanishWelfare")

Format

A data frame with 180 observations and 5 variables.

Freq

frequency.

Alcohol

factor indicating daily alcohol consumption: less than 1 unit (<1), 1-2 units (1-2) or more than 2 units (>2). 1 unit is approximately 1 bottle of beer or 4cl 40% alcohol.

Income

factor indicating income group in 1000 DKK (0-50, 50-100, 100-150, >150).

Status

factor indicating marriage status (Widow, Married, Unmarried).

Urban

factor indicating urbanization: Copenhagen (Copenhagen), Suburbian Copenhagen (SubCopenhagen), three largest cities (LargeCity), other cities (City), countryside (Country).

Source

E. B. Andersen (1991), The Statistical Analysis of Categorical Data, page 205.

References

E. B. Andersen (1991), The Statistical Analysis of Categorical Data. 2nd edition. Springer-Verlag, Berlin.

Examples

data("DanishWelfare")
ftable(xtabs(Freq ~ ., data = DanishWelfare))

Description

Diagnostic distribution plots: poissonness, binomialness and negative binomialness plots.

Usage

distplot(x, type = c("poisson", "binomial", "nbinomial"),
  size = NULL, lambda = NULL, legend = TRUE, xlim = NULL, ylim = NULL,
  conf_int = TRUE, conf_level = 0.95, main = NULL,
  xlab = "Number of occurrences", ylab = "Distribution metameter",
  gp = gpar(cex = 0.8), lwd=2, gp_conf_int = gpar(lty = 2),
  name = "distplot", newpage = TRUE,
  pop =TRUE, return_grob = FALSE, ...)

Arguments

Details

distplot plots the number of occurrences (counts) against the distribution metameter of the specified distribution. If the distribution fits the data, the plot should show a straight line. See Friendly (2000) for details.

In these plots, the open points show the observed count metameters; the filled points show the confidence interval centers, and the dashed lines show the conf_level confidence intervals for each point.

Value

Returns invisibly a data frame containing the counts (Counts), frequencies (Freq) and other details of the computations used to construct the plot.

Author(s)

Achim Zeileis [email protected]

References

D. C. Hoaglin (1980), A poissonness plot, The American Statistican, 34, 146–149.

D. C. Hoaglin & J. W. Tukey (1985), Checking the shape of discrete distributions. In D. C. Hoaglin, F. Mosteller, J. W. Tukey (eds.), Exploring Data Tables, Trends and Shapes, chapter 9. John Wiley & Sons, New York.

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

## Simulated data examples:
dummy <- rnbinom(1000, size = 1.5, prob = 0.8)
distplot(dummy, type = "nbinomial")

## Real data examples:
data("HorseKicks")
data("Federalist")
data("Saxony")
distplot(HorseKicks, type = "poisson")
distplot(HorseKicks, type = "poisson", lambda = 0.61)
distplot(Federalist, type = "poisson")
distplot(Federalist, type = "nbinomial", size = 1)
distplot(Federalist, type = "nbinomial")
distplot(Saxony, type = "binomial", size = 12)

Description

This function creates a doubledecker plot visualizing a classification rule.

Usage

## S3 method for class 'formula'
doubledecker(formula, data = NULL, ..., main = NULL)
## Default S3 method:
doubledecker(x, depvar = length(dim(x)),
  margins = c(1,4, length(dim(x)) + 1, 1),
  gp = gpar(fill = rev(gray.colors(tail(dim(x), 1)))),
  labeling = labeling_doubledecker,
  spacing = spacing_highlighting,
  main = NULL, keep_aspect_ratio = FALSE, ...)

Arguments

Details

Doubledecker plots visualize the the dependence of one categorical (typically binary) variable on further categorical variables. Formally, they are mosaic plots with vertical splits for all dimensions (antecedents) except the last one, which represents the dependent variable (consequent). The last variable is visualized by horizontal splits, no space between the tiles, and separate colors for the levels.

Value

The "structable" visualized is returned invisibly.

Author(s)

David Meyer [email protected]

References

H. Hoffmann (2001), Generalized odds ratios for visual modeling. Journal of Computational and Graphical Statistics, 10, 4, 628–640.

Meyer, D., Zeileis, A., and Hornik, K. (2006), The strucplot framework: Visualizing multi-way contingency tables with vcd. Journal of Statistical Software, 17(3), 1-48. doi:10.18637/jss.v017.i03 and available as vignette("strucplot").

See Also

strucplot, mosaic

Examples

data("Titanic")
doubledecker(Titanic)
doubledecker(Titanic, depvar = "Survived")
doubledecker(Survived ~ ., data = Titanic)

Description

Data from a 1974 Danish study given by Andersen (1991) on the employees who had been laid off. The workers are classified by their employment status on 1975-01-01, the cause of their layoff and the length of employment before they were laid off.

Usage

data("Employment")

Format

A 3-dimensional array resulting from cross-tabulating variables for 1314 employees. The variables and their levels are as follows:

Source

Michael Friendly (2000), Visualizing Categorical Data, pages 126–129.

References

E. B. Andersen (1991), The Statistical Analysis of Categorical Data. Springer-Verlag, Berlin.

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

data("Employment")

## Employment Status
mosaic(Employment,
       expected = ~ LayoffCause * EmploymentLength + EmploymentStatus,
       main = "Layoff*EmployLength + EmployStatus")

mosaic(Employment,
       expected = ~ LayoffCause * EmploymentLength + LayoffCause * EmploymentStatus,
       main = "Layoff*EmployLength + Layoff*EmployStatus")

## Stratified view

grid.newpage()
pushViewport(viewport(layout = grid.layout(ncol = 2)))
pushViewport(viewport(layout.pos.col = 1))

## Closure
mosaic(Employment[,,1], main = "Layoff: Closure", newpage = FALSE)

popViewport(1)
pushViewport(viewport(layout.pos.col = 2))

## Replaced
mosaic(Employment[,,2], main = "Layoff: Replaced", newpage = FALSE)
popViewport(2)

Description

Data from Mosteller & Wallace (1984) investigating the use of certain keywords (‘may’ in this data set) to identify the author of 12 disputed ‘Federalist Papers’ by Alexander Hamilton, John Jay and James Madison.

Usage

data("Federalist")

Format

A 1-way table giving the number of occurrences of ‘may’ in 262 blocks of text. The variable and its levels are

Source

Michael Friendly (2000), Visualizing Categorical Data, page 19.

References

F. Mosteller & D. L. Wallace (1984), Applied Bayesian and Classical Inference: The Case of the Federalist Papers. Springer-Verlag, New York, NY.

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

data("Federalist")
gf <- goodfit(Federalist, type = "nbinomial")
summary(gf)
plot(gf)

Description

Creates an (extended) fourfold display of a $2 \times 2 \times k$ contingency table, allowing for the visual inspection of the association between two dichotomous variables in one or several populations (strata).

Usage

fourfold(x,
  color = c("#99CCFF", "#6699CC", "#FFA0A0", "#A0A0FF", "#FF0000", "#000080"),
  conf_level = 0.95, std = c("margins", "ind.max", "all.max"),
  margin = c(1, 2), space = 0.2, main = NULL, sub = NULL,
  mfrow = NULL, mfcol = NULL, extended = TRUE, ticks = 0.15,
  p_adjust_method = p.adjust.methods, newpage = TRUE,
  fontsize = 12, default_prefix = c("Row", "Col", "Strata"),
  sep = ": ", varnames = TRUE, return_grob = FALSE)

Arguments

Details

The fourfold display is designed for the display of $2 \times 2 \times k$ tables.

Following suitable standardization, the cell frequencies $f_{ij}$ of each $2 \times 2$ table are shown as a quarter circle whose radius is proportional to $\sqrt{f_{ij}}$ so that its area is proportional to the cell frequency. An association (odds ratio different from 1) between the binary row and column variables is indicated by the tendency of diagonally opposite cells in one direction to differ in size from those in the other direction; color is used to show this direction. Confidence rings for the odds ratio allow a visual test of the null of no association; the rings for adjacent quadrants overlap iff the observed counts are consistent with the null hypothesis.

Typically, the number $k$ corresponds to the number of levels of a stratifying variable, and it is of interest to see whether the association is homogeneous across strata. The fourfold display visualizes the pattern of association. Note that the confidence rings for the individual odds ratios are not adjusted for multiple testing.

References

Friendly, M. (1994), A fourfold display for 2 by 2 by $k$ tables. Technical Report 217, York University, Psychology Department, http://datavis.ca/papers/4fold/4fold.pdf.

Friendly, M. (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

See Also

mosaic, assoc

link[stats]{p.adjust} for methods of p value adjustment

Examples

data("UCBAdmissions")
## Use the Berkeley admission data as in Friendly (1995).
x <- aperm(UCBAdmissions, c(2, 1, 3))
dimnames(x)[[2]] <- c("Yes", "No")
names(dimnames(x)) <- c("Sex", "Admit?", "Department")
ftable(x)

## Fourfold display of data aggregated over departments, with
## frequencies standardized to equate the margins for admission
## and sex.
## Figure 1 in Friendly (1994).
fourfold(margin.table(x, c(1, 2)))

## Fourfold display of x, with frequencies in each table
## standardized to equate the margins for admission and sex.
## Figure 2 in Friendly (1994).
fourfold(x)
cotabplot(x, panel = cotab_fourfold)

## Fourfold display of x, with frequencies in each table
## standardized to equate the margins for admission. but not
## for sex.
## Figure 3 in Friendly (1994).
fourfold(x, margin = 2)

Description

Fits a discrete (count data) distribution for goodness-of-fit tests.

Usage

goodfit(x, type = c("poisson", "binomial", "nbinomial"),
  method = c("ML", "MinChisq"), par = NULL)
## S3 method for class 'goodfit'
predict(object, newcount = NULL, type = c("response", "prob"), ...)
## S3 method for class 'goodfit'
residuals(object, type = c("pearson", "deviance",
"raw"), ...)
## S3 method for class 'goodfit'
print(x, residuals_type = c("pearson", "deviance",
"raw"), ...)

Arguments

Details

goodfit essentially computes the fitted values of a discrete distribution (either Poisson, binomial or negative binomial) to the count data given in x. If the parameters are not specified they are estimated either by ML or Minimum Chi-squared.

To fix parameters, par should be a named list specifying the parameters lambda for "poisson" and prob and size for "binomial" or "nbinomial", respectively. If for "binomial", size is not specified it is not estimated but taken as the maximum count.

The corresponding Pearson Chi-squared or likelihood ratio statistic, respectively, is computed and given with their $p$ values by the summary method. The summary method always prints this information and returns a matrix with the printed information invisibly. The plot method produces a rootogram of the observed and fitted values.

In case of count distribtions (Poisson and negative binomial), the minimum Chi-squared approach is somewhat ad hoc. Strictly speaking, the Chi-squared asymptotics would only hold if the number of cells were fixed or did not increase too quickly with the sample size. However, in goodfit the number of cells is data-driven: Each count is a cell of its own. All counts larger than the maximal count are merged into the cell with the last count for computing the test statistic.

Value

A list of class "goodfit" with elements:

Author(s)

Achim Zeileis [email protected]

References

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

See Also

rootogram

Examples

## Simulated data examples:
dummy <- rnbinom(200, size = 1.5, prob = 0.8)
gf <- goodfit(dummy, type = "nbinomial", method = "MinChisq")
summary(gf)
plot(gf)

dummy <- rbinom(100, size = 6, prob = 0.5)
gf1 <- goodfit(dummy, type = "binomial", par = list(size = 6))
gf2 <- goodfit(dummy, type = "binomial", par = list(prob = 0.6, size = 6))
summary(gf1)
plot(gf1)
summary(gf2)
plot(gf2)

## Real data examples:
data("HorseKicks")
HK.fit <- goodfit(HorseKicks)
summary(HK.fit)
plot(HK.fit)

data("Federalist")
## try geometric and full negative binomial distribution
F.fit <- goodfit(Federalist, type = "nbinomial", par = list(size = 1))
F.fit2 <- goodfit(Federalist, type = "nbinomial")
summary(F.fit)
summary(F.fit2)
plot(F.fit)
plot(F.fit2)

Description

Bar plots of 1-way tables in grid.

Usage

grid_barplot(height, width = 0.8, offset = 0,
  names = NULL, xlim = NULL, ylim = NULL, xlab = "", ylab = "", main = "",
  gp = gpar(fill = "lightgray"), name = "grid_barplot",
  newpage = TRUE, pop = FALSE, return_grob = FALSE)

Arguments

Details

grid_barplot mimics (some of) the features of barplot, but currently it only supports 1-way tables.

Author(s)

Achim Zeileis [email protected]

Examples

grid_barplot(sample(1:6), names = letters[1:6])

Description

This function can be used to add legends to grid-based plots.

Usage

grid_legend(x, y, pch = NA, col = par('col'), labels, frame = TRUE,
            hgap = unit(0.8, "lines"), vgap = unit(0.8, "lines"),
            default_units = "lines", gp = gpar(), draw = TRUE,
            title = NULL, just = 'center', lwd = NA, lty = NA,
            size = 1,
            gp_title = NULL, gp_labels = NULL,
            gp_frame = gpar(fill = "transparent"),
            inset = c(0, 0))

Arguments

Value

Invisibly, the legend as a "grob" object.

Author(s)

David Meyer [email protected] Florian Gerber [email protected]

See Also

legend

Examples

data("Lifeboats")
attach(Lifeboats)
ternaryplot(Lifeboats[,4:6],
  pch = ifelse(side == "Port", 1, 19),
  col = ifelse(side == "Port", "red", "blue"),
  id  = ifelse(men / total > 0.1, as.character(boat), NA),
  prop_size = 2,
  dimnames_position = "edge",
  main = "Lifeboats on Titanic")
grid_legend(0.8, 0.9, c(1, 19), c("red", "blue"),
  c("Port", "Starboard"), title = "SIDE")


grid.newpage()
pushViewport(viewport(height = .9, width = .9 ))
grid.rect(gp = gpar(lwd = 2, lty = 2))

grid_legend(x = unit(.05,'npc'),
            y = unit(.05,'npc'),
            just = c(0,0),
            pch = c(1,2,3),
            col = c(1,2,3),
            lwd=NA, 
            lty=NA,
            labels = c("b",'r','g'),
            title = NULL,
            gp=gpar(lwd=2, cex=1),
            hgap = unit(.8, "lines"),
            vgap = unit(.9, "lines"))

grid_legend(x = unit(1,'npc'),
            y = unit(1,'npc'),
            just = c(1,1),
            pch = NA,
            col = c(1,2,3,4),
            lwd=c(1,1,1,3), 
            lty=c(1,2,1,3),
            labels = c("black",'red','green','blue'),
            gp_labels = list(gpar(col = 1), gpar(col = 2), gpar(col = 3), gpar(col = 4)),
            title = NULL,
            gp=gpar(lwd=2, cex=1),
            hgap = unit(.8, "lines"),
            vgap = unit(.9, "lines"))

grid_legend(x = 'topleft',
            pch = c(1,NA,2,NA),
            col = c(1,2,3,4),
            lwd=NA, 
            lty=c(NA,2,NA,3),
            labels = c("black",'red','green','blue'),
            title = 'Some LONG Title',
            gp_title = gpar(col = 3),
            gp_frame = gpar(col = 4, lty = 2, fill = "transparent"),
            gp_labels = gpar(col = 6),
            gp=gpar(lwd=2, cex=2, col = 1),
            hgap = unit(.8, "lines"),
            vgap = unit(.9, "lines"))

grid_legend(x = .7,
            y = .7,
            pch = c(1,NA,2,NA),
            col = c(1,2,3,4),
            lwd=1, 
            lty=c(NA,2,NA,3),
            labels = c("black",'red','green','blue'),
            title = 'short T',
            gp=gpar(lwd=1, cex=.7,col = 1),
            hgap = unit(.8, "lines"),
            vgap = unit(.9, "lines"))

grid_legend(x = 'bottomright',
            pch = c(1,NA,2,NA),
            col = c(2),
            lwd=NA, 
            lty=c(NA,2,NA,3),
            labels = c("black",'red','green','blue'),
            title = NULL,
            gp=gpar(lwd=2, cex=1,col = 1),
            hgap = unit(.8, "lines"),
            vgap = unit(.9, "lines"))

Description

This data set is deduced from the Baseball fielding data set: fielding performance basically includes the numbers of Errors, Putouts and Assists made by each player. In order to reduce the number of observations, the was compressed by calculating the mean number of errors, putouts and assists for each team and for only 6 positions (1B, 2B, 3B, C, OF, SS and UT). In addition, each of these three variables was scaled to a common range by dividing each variable by the maximum of the variable.

Usage

data("Hitters")

Format

A data frame with 154 observations and 4 variables.

Positions

factor indicating the field position (1B=first baseman, 2B=second baseman, 3B=third baseman, C=catcher, OF=outfielder, SS=Short Stop, UT=Utility Players).

Putouts

occur when a fielder causes an opposing player to be tagged or forced out.

Assists

are credited to other fielders involved in making that putout.

Errors

count the errors made by a player.

Source

SAS System for Statistical Graphics, First Edition, Page A2.3

References

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

Examples

data("Hitters")
attach(Hitters)

colors <- c("black","red","green","blue","red","black","blue")
pch <- substr(levels(Positions), 1, 1)
ternaryplot(Hitters[,2:4],
  pch = as.character(Positions),
  col = colors[as.numeric(Positions)],
  main = "Baseball Hitters Data")
grid_legend(0.8, 0.9, pch, colors, levels(Positions),
  title = "POSITION(S)")

detach(Hitters)

Description

Create a HLS color from specifying hue, luminance and saturation.

Usage

hls(h = 1, l = 0.5, s = 1)

Arguments

Details

HLS colors are a similar specification of colors as HSV colors, but using hue/luminance/saturation rather that hue/saturation/value.

Author(s)

Achim Zeileis [email protected]

See Also

hsv, hcl2hex, polarLUV

Examples

## an HLS color wheel
pie(rep(1, 12), col = sapply(1:12/12, function(x) hls(x)))

Description

Data from von Bortkiewicz (1898), given by Andrews & Herzberg (1985), on number of deaths by horse or mule kicks in 10 (of 14 reported) corps of the Prussian army. 4 corps were not considered by Fisher (1925) as they had a different organization. This data set is a popular subset of the VonBort data.

Usage

data("HorseKicks")

Format

A 1-way table giving the number of deaths in 200 corps-years. The variable and its levels are

Source

Michael Friendly (2000), Visualizing Categorical Data, page 18.

References

D. F. Andrews & A. M. Herzberg (1985), Data: A Collection of Problems from Many Fields for the Student and Research Worker. Springer-Verlag, New York, NY.

R. A. Fisher (1925), Statistical Methods for Research Workers. Oliver & Boyd, London.

L. von Bortkiewicz (1898), Das Gesetz der kleinen Zahlen. Teubner, Leipzig.

M. Friendly (2000), Visualizing Categorical Data. SAS Institute, Cary, NC.

See Also

VonBort

Examples

data("HorseKicks")
gf <- goodfit(HorseKicks)
summary(gf)
plot(gf)

Description

The table relates the length of stay (in years) of 132 long-term schizophrenic patients in two London mental hospitals with the frequency of visits.

Usage

data("Hospital")

Format

A 2-dimensional array resulting from cross-tabulating 132 patients. The variables and their levels are as follows:

Details

Wing (1962) who collected this data concludes that the longer the length of stay in hospital, the less frequent the visits.

Haberman (1974) notes that this pattern does not increase from the "Less than monthly" to the "Never" group, which are homogeneous.

Source

S.J Haberman (1974): Log-linear models for frequency tables with ordered classifications. Biometrics, 30:689–700.

References

J.K. Wing (1962): Institutionalism in mental hospitals. British Journal of Social Clinical Psychology, 1:38–51.

Examples

data("Hospital")

mosaic(t(Hospital), shade = TRUE)
mosaic(Hospital, shade = TRUE)
sieve(Hospital, shade = TRUE)
assoc(Hospital, shade = TRUE)

Description

Computes table of expected frequencies (under the null hypotheses of independence) from an $n$-way table.

Usage

independence_table(x, frequency = c("absolute", "relative"))

Arguments

Value

A table with either absolute or relative frequencies.

Author(s)

David Meyer [email protected]

Examples

data("MSPatients")
independence_table(MSPatients)
independence_table(MSPatients, frequency = "relative")

Description

Data from Petersen (1968) about the job satisfaction of 715 blue collar workers, selected from Danish Industry in 1968.

Usage

data("JobSatisfaction")

Format

A data frame with 8 observations and 4 variables.

Freq

frequency.

management

factor indicating quality of management (bad, good).

supervisor

factor indicating supervisor's job satisfaction (low, high).

own

factor indicating worker's own job satisfaction (low, high).

Source

E. B. Andersen (1991), The Statistical Analysis of Categorical Data, Table 5.4.

References

E. B. Andersen (1991), The Statistical Analysis of Categorical Data. 2nd edition. Springer-Verlag, Berlin.

E. Petersen (1968), Job Satisfaction in Denmark. (In Danish). Mentalhygiejnisk Forlag, Copenhagen.

Examples

data("JobSatisfaction")
structable(~ ., data = JobSatisfaction)

mantelhaen.test(xtabs(Freq ~ own + supervisor + management,
                      data = JobSatisfaction))

Description

Data from a Danish study in 1983 and 1985 about sports activities and the opinion about joint sports with the other gender among 16–19 year old high school students.

Usage

data("JointSports")

Format

A data frame with 40 observations and 5 variables.

Freq

frequency.

opinion

factor indicating opinion about sports joint with the other gender (very good, good, indifferent, bad, very bad).

year

factor indicating year of study (1983, 1985).

grade

factor indicating school grade (1st, 3rd).

gender

factor indicating gender (Boy, Girl).

Source

E. B. Andersen (1991), The Statistical Analysis of Categorical Data, page 210.

References

E. B. Andersen (1991), The Statistical Analysis of Categorical Data. 2nd edition. Springer-Verlag, Berlin.

Examples

library(MASS)
data("JointSports")
tab <- xtabs(Freq ~ gender + opinion + grade + year, data = JointSports)
doubledecker(opinion ~ gender + year + grade, data = tab)
loglm(~ opinion* (gender + grade+ year) + gender*year*grade, data = tab)

Description

Computes two agreement rates: Cohen's kappa and weighted kappa, and confidence bands.

Usage

Kappa(x, weights = c("Equal-Spacing", "Fleiss-Cohen"))
## S3 method for class 'Kappa'
print(x, digits=max(getOption("digits") - 3, 3),
CI=FALSE, level=0.95, ...)
## S3 method for class 'Kappa'
confint(object, parm, level = 0.95, ...)
## S3 method for class 'Kappa'
summary(object, ...)
## S3 method for class 'summary.Kappa'
print(x, ...)

Arguments

Details

Cohen's kappa is the diagonal sum of the (possibly weighted) relative frequencies, corrected for expected values and standardized by its maximum value. The equal-spacing weights are defined by $1 - |i - j| / (r - 1)$, $r$ number of columns/rows, and the Fleiss-Cohen weights by $1 - |i - j|^2 / (r - 1)^2$. The latter one attaches greater importance to near disagreements.

Value

An object of class "Kappa" with three components:

Note

The summary method also prints the weights.

There is a confint method for computing approximate confidence intervals.

Author(s)

David Meyer [email protected]

References

Cohen, J. (1960), A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20, 37–46.

Everitt, B.S. (1968), Moments of statistics kappa and weighted kappa. The British Journal of Mathematical and Statistical Psychology, 21, 97–103.

Fleiss, J.L., Cohen, J., and Everitt, B.S. (1969), Large sample standard errors of kappa and weighted kappa. Psychological Bulletin, 72, 332–327.

See Also

agreementplot, confint

Examples

data("SexualFun")
K <- Kappa(SexualFun)
K
confint(K)
summary(K)
print(K, CI = TRUE)

Description

These functions generate labeling functions used for strucplots.

Usage

labeling_border(labels = TRUE, varnames = labels,
  set_labels = NULL, set_varnames = NULL, 
  tl_labels = NULL, alternate_labels = FALSE, tl_varnames = NULL, 
  gp_labels = gpar(fontsize = 12),
  gp_varnames = gpar(fontsize = 12, fontface = 2),
  rot_labels = c(0, 90, 0, 90), rot_varnames = c(0, 90, 0, 90),
  pos_labels = "center", pos_varnames = "center",
  just_labels = "center", just_varnames = pos_varnames,
  boxes = FALSE, fill_boxes = FALSE,
  offset_labels = c(0, 0, 0, 0), offset_varnames = offset_labels,
  labbl_varnames = NULL, labels_varnames = FALSE, sep = ": ",
  abbreviate_labs = FALSE, rep = TRUE, clip = FALSE, ...)
labeling_values(value_type = c("observed", "expected", "residuals"),
                suppress = NULL, digits = 1, clip_cells = FALSE, ...)
labeling_residuals(suppress = NULL, digits = 1, clip_cells = FALSE, ...)
labeling_conditional(...)
labeling_left(rep = FALSE, pos_varnames = "left",
  pos_labels = "left", just_labels = "left", ...)
labeling_left2(tl_labels = TRUE, clip = TRUE, pos_varnames = "left",
  pos_labels = "left", just_labels = "left", ...)
labeling_cboxed(tl_labels = TRUE, boxes = TRUE, clip = TRUE,
  pos_labels = "center", ...)
labeling_lboxed(tl_labels = FALSE, boxes = TRUE, clip = TRUE,
  pos_labels = "left", just_labels = "left",
  labbl_varnames = FALSE, ...)
labeling_doubledecker(lab_pos = c("bottom", "top"), dep_varname = TRUE,
  boxes = NULL, clip = NULL, labbl_varnames = FALSE,
  rot_labels = rep.int(0, 4),
  pos_labels = c("left", "center", "left", "center"),
  just_labels = c("left", "left", "left", "center"),
  varnames = NULL, gp_varnames = gpar(fontsize = 12, fontface = 2),
  offset_varnames = c(0, -0.6, 0, 0), tl_labels = NULL, ...)

Arguments

Details

These functions generate labeling functions called by strucplot for their side-effect of adding labels to the plot. They suppose that a strucplot has been drawn and the corresponding viewport structure is pushed, since the positions of the viewports are used for the label positioning. Note that the functions can also be used ‘stand-alone’ as shown in the examples.

All values supplied to vectorized arguments can be ‘abbreviated’ by using named components which override the default component values. In addition, these defaults can be overloaded by the sequence of non-named components which are recycled as needed (see examples).

This help page only documents labeling_border and derived functions, more functions are described on the help page for labeling_cells and labeling_list.

labeling_left, labeling_left2, labeling_cboxed, and labeling_lboxed are really just wrappers to labeling_border, and good examples for the parameter usage.

labeling_residuals is a trivial wrapper for labeling_values, which in turn calls labeling_border by additionally adding the observed or expected frequencies or residuals to the cells.

Value

A function with arguments:

Author(s)

David Meyer [email protected]

References

Meyer, D., Zeileis, A., and Hornik, K. (2006), The strucplot framework: Visualizing multi-way contingency tables with vcd. Journal of Statistical Software, 17(3), 1-48. doi:10.18637/jss.v017.i03 and available as vignette("strucplot").

See Also

labeling_cells, labeling_list, structable, grid.text

Examples

data("Titanic")

mosaic(Titanic)

mosaic(Titanic, labeling = labeling_left)
labeling_left

mosaic(Titanic, labeling = labeling_cboxed)
labeling_cboxed

mosaic(Titanic, labeling = labeling_lboxed)
labeling_lboxed

data("PreSex")
mosaic(~ PremaritalSex + ExtramaritalSex | Gender + MaritalStatus,
  data = PreSex, labeling = labeling_conditional)

## specification of vectorized arguments 
mosaic(Titanic, labeling_args = list(abbreviate_labs = c(Survived = TRUE)))

mosaic(Titanic, labeling_args = list(clip = TRUE, boxes = TRUE,
  fill_boxes = c(Survived = "green", "red")))

mosaic(Titanic, labeling_args = list(clip = TRUE, boxes = TRUE,
  fill_boxes = list(Sex = "red", "green")))

mosaic(Titanic, labeling_args = list(clip = TRUE, boxes = TRUE,
  fill_boxes = list(Sex = c(Male = "red", "blue"), "green")))

## change variable names
mosaic(Titanic, labeling_args = list(set_varnames = c(Sex = "Gender")))

## change labels
mosaic(Titanic, labeling_args = list(set_varnames = c(Survived = "Status"),
  set_labels = list(Survived = c("Survived", "Not Survived")), rep = FALSE))

## show frequencies
mosaic(Titanic, labeling = labeling_values)

Description

These functions generate labeling functions that produce labels for strucplots.

Usage

labeling_cells(labels = TRUE, varnames = TRUE,
  abbreviate_labels = FALSE, abbreviate_varnames = FALSE,
  gp_text = gpar(), lsep = ": ", lcollapse = "\n",
  just = "center", pos = "center", rot = 0,
  margin = unit(0.5, "lines"), clip_cells = TRUE,
  text = NULL, ...)
labeling_list(gp_text = gpar(), just = "left", pos = "left", lsep = ": ",
  sep = " ", offset = unit(c(2, 2), "lines"),
  varnames = TRUE, cols = 2, ...)

Arguments