Package 'factorplot'

Title: Presenting Pairwise Comparisons
Description: The tools herein calculate, print, summarize and plot pairwise differences that result from generalized linear models, general linear hypothesis tests and multinomial logistic regression models. For more information, see Armstrong (2013) <doi:10.32614/RJ-2013-021>.
Authors: Dave Armstrong [aut, cre]
Maintainer: Dave Armstrong <[email protected]>
License: GPL (>= 2)
Version: 1.2.3
Built: 2024-11-20 06:39:58 UTC
Source: CRAN

Help Index


Calculate Pairwise Differences

Description

This function calculates all pairwise difference from the input data. The input data can be the result of a GLM (produced with glm), a multinomial logit model (produced with multinom from the nnet package), a general linear hypothesis test (produced with glht from the multcomp package), an object of class eff from the effects package or any vector of values and a corresponding variance-covariance matrix.

Usage

factorplot(obj, adjust.method = "none", ...)

## S3 method for class 'glm'
factorplot(
  obj,
  adjust.method = "none",
  order = "natural",
  factor.variable = NULL,
  pval = 0.05,
  two.sided = TRUE,
  ...
)

## S3 method for class 'lm'
factorplot(
  obj,
  adjust.method = "none",
  order = "natural",
  factor.variable = NULL,
  pval = 0.05,
  two.sided = TRUE,
  ...
)

## S3 method for class 'summary.glht'
factorplot(obj, ...)

## S3 method for class 'glht'
factorplot(obj, adjust.method = "none", pval = 0.05, ...)

## S3 method for class 'sims'
factorplot(obj, adjust.method = "none", order = "natural", pval = 0.05, ...)

## Default S3 method:
factorplot(
  obj,
  adjust.method = "none",
  order = "natural",
  var,
  resdf = Inf,
  pval = 0.05,
  two.sided = TRUE,
  ...
)

## S3 method for class 'eff'
factorplot(
  obj,
  adjust.method = "none",
  order = "natural",
  pval = 0.05,
  two.sided = TRUE,
  ordby = NULL,
  ...
)

## S3 method for class 'multinom'
factorplot(
  obj,
  adjust.method = "none",
  order = "natural",
  variable,
  pval = 0.05,
  two.sided = TRUE,
  ...
)

Arguments

obj

An object of class glm or lm, glht, summary.glht, multinom or a vector of values (of class numeric) for which pairwise differences will be calculated. Alternatively, an object of class sims which gives an Nsim x Nstimulus matrix of predictions from which differences will be calculated.

adjust.method

For objects of class multinom and numeric - one of the methods allowed by p.adjust in stats - ‘holm’, ‘hochberg’, ‘hommel’, ‘bonferroni’, ‘BH’, ‘BY’, ‘fdr’, ‘none’. See help for the p.adjust for more information on these different adjustment methods. For objects of class glm, lm or glht, additional arguments of ‘single-step’, ‘Shaffer’, ‘Westfall’ and ‘free’ are possible. See glht from the multcomp package for details.

...

Additional arguments to be passed to summary.glht, including, but not limited to level and alternative.

order

One of ‘natural’, ‘alph’, or ‘size’ indicating how the levels of the factor should be ordered for presentation. The ‘natural’ option (the default) leaves the levels as they are in the factor contrasts. ‘alph’ sorts the levels alphabetically and ‘size’ sorts the levels by size of coefficient.

factor.variable

String containing the name of the factor for which pairwise coefficient differences will be calculated (if a glm or lm class object is passed to the function)

pval

The (uncorrected) Type I error probability required, default = 0.05

two.sided

Logical argument indicating whether the hypothesis test should be against a two-sided alternative if TRUE (default) or a one-sided alternative if FALSE

var

Variance-covariance matrix to be used if obj is a numeric vector. This could also be a vector of quasi/floating variances from which a diagonal variance-covariance matrix will be produced

resdf

Residual degrees of freedom used as the degrees of freedom for the t-distribution from which p-values will be generated if obj is a numeric vector

ordby

For objects of class eff with interactions, ordby is a string indicating the variable by which the plot should be ordered.

variable

String containing the name of the column of the model matrix for which pairwise differences will be calculated if a multinom class object is passed to the function

Details

This function calculates pairwise differences that can be passed to a novel plotting method that does not suffer from some of the same problems as floating/quasi confidence intervals and is easier to apprehend immediately than a compact letter display.

While the factorplot function and its print and summary methods work equally well regardless of the number of levels in the factor.variable, the plot function automatically scales the resulting graph to the appropriate size, but will be less useful as the number of contrasts gets large (e.g., > 30). If more than one factor covariate is present and the factor.variable option is NULL, the function generates a text-based menu in the R GUI that will allow the users to pick the term for which they want to calculate the results.

Value

b.diff

An upper-triangular matrix of pairwise differences between row and column levels of the factor

b.sd

An upper-triangular matrix of standard errors of the pairwise differences represented in b.diff

pval

An upper-triangular matrix of uncorrected (one-sided) p-values corresponding to the entries of b.diff

p

The p-value specified in the command

Author(s)

Dave Armstrong

References

Easton, D.F., J. Peto and G.A.G. Babiker. 1991. Floating absolute risk: An alternative to relative risk in survival and case control analysis avoiding an arbitrary reference group. Statistics in Medicine 10: 1025–1035.
Firth, David and Renee X. de Menzes. 2004. Quasi-variances. Biometrika 91.1: 65–80.
Plummer, M. 2004. Improved estimates of floating absolute risk. Statistics in Medicine 23: 93–104.

Examples

## for lm/glm
x <- as.factor(round(runif(1000, .5,5.5)))
levels(x) <- paste("lab", 1:20, sep="")
X <- model.matrix(~x)
Y <- X %*% rnorm(ncol(X),0,4) + rnorm(1000)
mod <- lm(Y ~ x)
fp <- factorplot(mod, factor.variable="x",  pval = 0.05, order="alph")

## for glht
library(multcomp)
mod.glht <- glht(mod, linfct = mcp('x' = 'Tukey'))
fp2 <- factorplot(mod.glht, adjust.method='single-step')

## for vector of values
b <- c(0, mod$coef[-1])
v <- rbind(0, cbind(0, vcov(mod)[-1,-1]))
names(b) <- colnames(v) <- rownames(v) <- mod$xlevels[["x"]]
fp3 <- factorplot(b, var=v, resdf=mod$df.residual)

## for multinomial logit
data(france)
library(nnet)
multi.mod <- multinom(vote ~ retnat + lrself + male + age, data=france)
fp4 <- factorplot(multi.mod, variable="lrself")

Example data for factorplot function

Description

A subset of data from the 1994 Eurobarometer for France

Format

A data frame with 542 observations on the following 5 variables.

lrself

respondent's left-right self-placement on a 1(left)-10(right) scale

male

a dummy variable coded 1 for males and 0 for females

age

respondent's age

vote

a factor indicating vote choice with levels PCF, PS, Green, RPR and UDF

retnat

a factor indicating the respondent's retrospective national economic evaluation with levels Better, Same and Worse

References

Reif, Karlheinz and Eric Marlier. 1997. Euro-barometer 42.0: The First Year of the New European Union, November-December 1994. Inter-university Consortium for Political and Social Research (ICPSR) [distributor].


Plot method for objects of class factorplot

Description

Creates a plot akin to an upper-triangular levelplot (though using plot rather than levelplot) where the coloring of the squares represents significance and text inside the squares represents the pairwise difference and its correspopnding standard error.

Usage

## S3 method for class 'factorplot'
plot(
  x,
  ...,
  abbrev.char = 10,
  polycol = NULL,
  textcol = NULL,
  trans = NULL,
  print.sig.leg = TRUE,
  print.square.leg = TRUE,
  scale.text = 1,
  space.text = 1,
  print.est = TRUE,
  print.se = TRUE
)

Arguments

x

An object of class factorplot, produced by factorplot.

...

Other arguments to be passed to plot, currently not implemented

abbrev.char

The number of characters that should be used to abbreviate the levels of the factor. Set to a large value for unabbreviated names.

polycol

A vector of three colors indicating the colors of polygons when the difference is significant negative, insignificant, and significant positive, in that order. Defaults to c(‘gray80’, ‘white’, ‘gray40’).

textcol

A vector of three colors indicating the text color for polygons that are significant negative, insignificant, and significant positive, in that order. Defaults to c(‘black’, ‘black’, ‘white’)

trans

A character string representing the post-hypothesis-testing transformation to be performed on the estimates. For example, if the estimates provided to the factorplot command are log-floating absolute risks, you could use the transformation ‘exp’. The transformation is performed through a call to do.call

print.sig.leg

logical indicating whether the legend identifying the meaning of the different colors should be included.

print.square.leg

logical indicating whether the legend identifying the meaning of the numbers in each square should be included.

scale.text

optional scale factor to be applied to text, numbers bigger than 1 make text bigger than default and numbers smaller than 1 do the opposite

space.text

optional text spacing factor, numbers bigger than 1 push text toward the extent of the boxes and numbers smaller than one bring text in toward the center

print.est

logical argument indicating whether the estimates should be printed in the boxes

print.se

logical argument indicating whether the standard errors should be printed in the boxes

Value

a graph

For m categories, the plot returns an m-1 x m-1 matrix where the nexus of the row and column values represent the pairwise differencee between the row and column values along with the standard error of the difference on the linear scale (unless a transformation is performed).

Author(s)

Dave Armstrong

See Also

factorplot

Examples

est1 <- log(c(1.00,2.12,1.44,1.31,1.44,
     1.46,0.90))
var1 <- c(0.242,0.096,0.156,0.140,
     0.380,0.484,0.375)^2
names(est1) <- c(
     "Normal & superficial gastritis", 
     "Chronic gastritis", 
     "Chronic atrophic gastritits", 
     "Intestinal metaplasia I", 
     "Intestinal metaplasia II", 
     "Intestinal metaplasia III", 
     "Dysplasia")

plummer_fp1 <- factorplot(est1, var=var1, resdf=Inf)
plot(plummer_fp1, trans="exp", abbrev.char = 100)

Print method for objects of class factorplot

Description

Prints the output from an object of class factorplot. By default, the function prints all pairwise differences along with standard errors and p-values (optionally adjusted for multiple testing). Optionally, it can print only significant differences.

Usage

## S3 method for class 'factorplot'
print(x, ..., digits = 3, sig = FALSE, trans = NULL)

Arguments

x

An object of class factorplot.

...

Other arguments passed to print, currently not implemented

digits

The number of digits to print in each column

sig

Logical indicating whether only significant differences should be printed.

trans

A character string representing the post-hypothesis-testing transformation to be performed on the estimates. For example, if the estimates provided to the factorplot command are log-floating absolute risks, you could use the transformation ‘exp’. The transformation is performed through a call to do.call

Value

Printed output

The printed output shows the difference between all pairs of stimuli (i.e., levels of the factor) along with their standard errors and (optionally adjusted) p-values. If a transformation is implemented, the difference is transformed accordingly, but the standard errors and other values are on the linear scale.

Author(s)

Dave Armstrong

See Also

factorplot

Examples

est1 <- log(c(1.00,2.12,1.44,1.31,1.44,
     1.46,0.90))
var1 <- c(0.242,0.096,0.156,0.140,
     0.380,0.484,0.375)^2
names(est1) <- c(
     "Normal & superficial gastritis", 
     "Chronic gastritis", 
     "Chronic atrophic gastritits", 
     "Intestinal metaplasia I", 
     "Intestinal metaplasia II", 
     "Intestinal metaplasia III", 
     "Dysplasia")
plummer_fp1 <- factorplot(est1, var=var1, resdf=Inf)
print(plummer_fp1, trans="exp")

Auxiliary Function to Plot a Square

Description

An auxiliary function to plot squares, used by the plot.factorplot function

Usage

squares(ll, width = 1, col)

Arguments

ll

The (x,y) coordinate of the lower-left corder of the square

width

a scalar indicating how wide the squares should be

col

a color with which the square will be filled in

Details

This is a function called by plot.factorplot and not intended to be directly used by the user; however, it is possible that this could be of more general use as a utility. The function is simply a wrapper to polygon that obviates the need to specify all (x,y) coordinates for the polygon.

Value

square

A square is printed on the graph, but nothing else is returned

Author(s)

Dave Armstrong


Summary method for objects of class factorplot

Description

Summarizes the number of significant positive and negative differences for objects of class factorplot

Usage

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

Arguments

object

An object of class factorplot

...

Other arguments passed to summary, currently not implemented

Value

Printed Output

The printed output summarises the number of stimuli that are significantly higher or lower and not significantly different from each other.

Author(s)

Dave Armstrong

See Also

factorplot

Examples

x <- as.factor(round(runif(1000, .5,5.5)))
levels(x) <- paste("lab", 1:20, sep="")
X <- model.matrix(~x)
b <- rnorm(ncol(X),0,4)
Y.hat <- X %*% b 
Y <- Y.hat  + rnorm(1000)
mod <- lm(Y ~ x)
fp <- factorplot(mod, factor.variable="x", pval=0.05, order="alph")
summary(fp)