Package 'relimp' reference manual

Title:	Relative Contribution of Effects in a Regression Model
Description:	Functions to facilitate inference on the relative importance of predictors in a linear or generalized linear model, and a couple of useful Tcl/Tk widgets.
Authors:	David Firth with contributions from Heather Turner and John Fox
Maintainer:	David Firth <[email protected]>
License:	GPL (>= 2)
Version:	1.0-5
Built:	2024-11-06 06:39:42 UTC
Source:	CRAN

Pick Subsets from a Vector

Description

Provides a Tk dialog or a text-based menu for interactive selection of one or more subsets from a vector.

Usage

pickFrom(vec, nsets = 1, return.indices = FALSE,
                 setlabels = NULL, edit.setlabels = TRUE,
                 subset = TRUE,
                 warningText = "one or more selections empty",
                 title = "Subset picker",
                 items.label = "Pick from",
                 labels.prompt = "Your label for this set",
                 list.height = 20,
                 items.scrollbar = TRUE,
                 preserve.order = TRUE,
                 graphics = TRUE,
                 listFont = "Courier 12",
                 labelFont = "Helvetica 11",
                 windowPos = "+150+30")
pickFrom(vec, nsets = 1, return.indices = FALSE,
                 setlabels = NULL, edit.setlabels = TRUE,
                 subset = TRUE,
                 warningText = "one or more selections empty",
                 title = "Subset picker",
                 items.label = "Pick from",
                 labels.prompt = "Your label for this set",
                 list.height = 20,
                 items.scrollbar = TRUE,
                 preserve.order = TRUE,
                 graphics = TRUE,
                 listFont = "Courier 12",
                 labelFont = "Helvetica 11",
                 windowPos = "+150+30")

Arguments

`vec`	a vector
`nsets`	a positive integer, the number of subsets to be selected
`return.indices`	logical, whether indices (`TRUE`) or vector contents (`FALSE`) are to be returned
`setlabels`	a character vector of labels for the subsets
`edit.setlabels`	logical, determines whether a textbox is provided for editing the label of each subset
`subset`	logical, character or numeric vector indicating which elements of `vec` should be made available for selection. Default is to make all elements available.
`warningText`	character, text to use as a warning in situations where no selection is made into one or more of the specified sets
`title`	character, title of the Tk dialog window
`items.label`	character, a label for the set of items to be selected from
`labels.prompt`	character, a prompt for textual set label(s)
`list.height`	maximum number of elements of `vec` to display at once
`items.scrollbar`	logical, whether a scrollbar is to be provided when `vec` is longer than `list.height`
`preserve.order`	logical: should the order of items in `vec` be maintained in all of the returned subsets?
`graphics`	logical: should a dialog be used, if possible?
`listFont`	a Tk font specification for the items list and subsets
`labelFont`	a Tk font specification for the labels entrybox
`windowPos`	position of the Tk dialog, in pixels from top left of display

Details

If graphics = TRUE and the tcltk package is operational, a Tk dialog is used, otherwise a text menu.

If return.indices is used together with subset, the indices returned relate to vec, not to vec[subset].

Value

EITHER (in the case of a text menu or if the dialog is ended with "OK") a list, with nsets components. Each component is a selected sub-vector, or a numeric vector of indices for a selected sub-vector (if return.indices is TRUE). The component names are as specified in setlabels, or as specified interactively.

OR (if the dialog is ended either "Cancel" or the close-window control button is used) NULL.

Author(s)

David Firth, with contributions from Heather Turner

Examples

## These examples cannot be run by example() but should be OK when pasted
## into an interactive R session
## Not run: 
pickFrom(c("apple", "banana", "plum", "grapefruit"), nsets = 2,
  preserve.order = FALSE,
  setlabels = c("Fruits I like", "Fruits I tolerate"))

## End(Not run)
## Not run: 
## Type selections as e.g. 1:2, 4
pickFrom(c("apple", "banana", "plum", "grapefruit"), nsets = 2,
  preserve.order = FALSE,
  setlabels = c("Fruits I like", "Fruits I tolerate"),
  graphics = FALSE)

## End(Not run)## These examples cannot be run by example() but should be OK when pasted
## into an interactive R session
## Not run: 
pickFrom(c("apple", "banana", "plum", "grapefruit"), nsets = 2,
  preserve.order = FALSE,
  setlabels = c("Fruits I like", "Fruits I tolerate"))

## End(Not run)
## Not run: 
## Type selections as e.g. 1:2, 4
pickFrom(c("apple", "banana", "plum", "grapefruit"), nsets = 2,
  preserve.order = FALSE,
  setlabels = c("Fruits I like", "Fruits I tolerate"),
  graphics = FALSE)

## End(Not run)

Convert a Character Vector to Tcl Format

Description

Converts a character vector into a brace-delimited Tcl list

Usage

R.to.Tcl(character.vector)
R.to.Tcl(character.vector)

Arguments

character.vector

A character vector

Value

A character vector of length 1

Author(s)

David Firth, [email protected]

Examples

R.to.Tcl(c("apple","banana"))
R.to.Tcl(c("apple","banana"))

Relative Importance of Predictors in a Regression Model

Description

Produces a summary of the relative importance of two predictors or two sets of predictors in a fitted model object.

Usage

relimp(object, set1=NULL,  set2=NULL, label1="set1", label2="set2", 
          subset=TRUE, 
          response.cat=NULL, ...)
## S3 method for class 'relimp'
print(x, digits=3, ...)
relimp(object, set1=NULL,  set2=NULL, label1="set1", label2="set2", 
          subset=TRUE, 
          response.cat=NULL, ...)
## S3 method for class 'relimp'
print(x, digits=3, ...)

Arguments

`object`	A model object of class `lm`, `glm`, `coxph`, `survreg`, `multinom`, `polr` or `gls`
`set1`	An index or vector of indices for the effects to be included in the numerator of the comparison
`set2`	An index or vector of indices for the effects to be included in the denominator of the comparison
`label1`	A character string; mnemonic name for the variables in `set1`
`label2`	A character string; mnemonic name for the variables in `set2`
`subset`	Either a vector of numeric indices for the cases to be included in the standardization of effects, or a vector of logicals (`TRUE` for inclusion) whose length is the same as the number of rows in the model frame, `object$model`. The default choice is to include all cases in the model frame.
`response.cat`	If `object` is of class `multinom`, this is a character string used to specify which regression is of interest (i.e., the regression which predicts the log odds on `response cat` versus the model's reference category). The `response.cat` argument should be an element of `object$lab`; or `NULL` if `object` is not of class `multinom`.
`...`	For models of class `glm`, one may additionally set the dispersion parameter for the family (for example, `dispersion=1.69`). By default it is obtained from `object`. Supplying it here permits explicit allowance for over-dispersion, for example.
`x`	an object of class `relimp`
`digits`	The number of decimal places to be used in the printed summary. Default is 3.

Details

If set1 and set2 both have length 1, relative importance is measured by the ratio of the two standardized coefficients. Equivalently this is the ratio of the standard deviations of the two contributions to the linear predictor, and this provides the generalization to comparing two sets rather than just a pair of predictors.

The computed ratio is the square root of the variance-ratio quantity denoted as ‘omega’ in Silber, J H, Rosenbaum, P R and Ross, R N (1995). Estimated standard errors are calculated by the delta method, as described in that paper for example.

If set1 and set2 are unspecified, and if the tcltk package has been loaded, a dialog box is provided (by a call to pickFrom) for the choice of set1 and set2 from the available model coefficients.

Value

An object of class relimp, with at least the following components:

`model`	The call used to construct the model object summarized
`sets`	The two sets of indices specified as arguments
`log.ratio`	The natural logarithm of the ratio of effect standard deviations corresponding to the two sets specified
`se.log.ratio`	An estimated standard error for log.ratio

If dispersion was supplied as an argument, its value is stored as the dispersion component of the resultant object.

Author(s)

David Firth [email protected]

References

Silber, J. H., Rosenbaum, P. R. and Ross, R N (1995) Comparing the Contributions of Groups of Predictors: Which Outcomes Vary with Hospital Rather than Patient Characteristics? JASA 90, 7–18.

Examples

set.seed(182)  ## an arbitrary number, just for reproducibility
x <- rnorm(100)
z <- rnorm(100)
w <- rnorm(100)
y <- 3 + (2 * x) + z + w + rnorm(100)
test <- lm(y ~ x + z + w)
print(test)
relimp(test, 2, 3)    #  compares effects of x and z
relimp(test, 2, 3:4)  #  compares effect of x with that of (z,w) combined
##
##  Data on housing and satisfaction, from Venables and Ripley
##  -- multinomial logit model
library(MASS)
library(nnet)
data(housing)
house.mult <- multinom(Sat ~ Infl + Type + Cont, weights = Freq,
  data = housing)
relimp(house.mult, set1 = 2:3, set2 = 7, response.cat = "High")
set.seed(182)  ## an arbitrary number, just for reproducibility
x <- rnorm(100)
z <- rnorm(100)
w <- rnorm(100)
y <- 3 + (2 * x) + z + w + rnorm(100)
test <- lm(y ~ x + z + w)
print(test)
relimp(test, 2, 3)    #  compares effects of x and z
relimp(test, 2, 3:4)  #  compares effect of x with that of (z,w) combined
##
##  Data on housing and satisfaction, from Venables and Ripley
##  -- multinomial logit model
library(MASS)
library(nnet)
data(housing)
house.mult <- multinom(Sat ~ Infl + Type + Cont, weights = Freq,
  data = housing)
relimp(house.mult, set1 = 2:3, set2 = 7, response.cat = "High")

Comparison of Relative Importances in a Multinomial Logit Model

Description

Produces a summary of the relative importance of two predictors or two sets of predictors in a fitted multinom model object, and compares relative importances across two of the fitted logit models.

Usage

relrelimp(object, set1=NULL, set2=NULL, label1="set1", label2="set2", 
          subset=TRUE, 
          response.cat1=NULL, response.cat2=NULL)
relrelimp(object, set1=NULL, set2=NULL, label1="set1", label2="set2", 
          subset=TRUE, 
          response.cat1=NULL, response.cat2=NULL)

Arguments

`object`	A model object of class `multinom`
`set1`	An index or vector of indices for the effects to be included in the numerator of the comparison
`set2`	An index or vector of indices for the effects to be included in the denominator of the comparison
`label1`	A character string; mnemonic name for the variables in `set1`
`label2`	A character string; mnemonic name for the variables in `set2`
`subset`	Either a vector of numeric indices for the cases to be included in the standardization of effects, or a vector of logicals (`TRUE` for inclusion) whose length is the same as the number of rows in the model frame, `object$model`. The default choice is to include all cases in the model frame.
`response.cat1`	A character string used to specify the first regression of interest (i.e., the regression which predicts the log odds on `response.cat1` versus the model's reference category). The `response.cat1` argument should be an element of `object$lab`.
`response.cat2`	A character string used to specify the second regression of interest (i.e., the regression which predicts the log odds on `response.cat2` versus the model's reference category). The `response.cat2` argument should be an element of `object$lab`.

Details

Computes a relative importance summary as described in relimp, for each of the two regressions specified by response.cat1 and response.cat2 (relative to the same reference category); and computes the difference of those two relative importance summaries, along with an estimated standard error for that difference.

Value

An object of class relrelimp, with at least the following components:

`model`	The call used to construct the model object summarized
`sets`	The two sets of indices specified as arguments
`response.category`	A character vector containing the specified `response.cat1` and `response.cat2`
`log.ratio`	The natural logarithm of the ratio of effect standard deviations corresponding to the two sets specified. A vector with three components: the first is for `response.cat1` versus the reference category, the second for `response.cat2` versus the reference category, the third is the difference.
`se.log.ratio`	Estimated standard errors for the elements of `log.ratio`

Author(s)

David Firth, [email protected]

Examples

##  Data on housing and satisfaction, from Venables and Ripley
library(MASS)
library(nnet)
data(housing)
house.mult <- multinom(Sat ~ Infl + Type + Cont, weights = Freq,
  data = housing)
relrelimp(house.mult, set1 = 2:3, set2 = 7, 
                      label1 = "Influence", label2 = "Contact",
                      response.cat1 = "Medium", response.cat2 = "High")
## Computes the relative contribution of Influence and Contact in 
## each of the two logistic regressions (Med/Low and High/Low), and
## compares those two relative-contribution measures.
##  Data on housing and satisfaction, from Venables and Ripley
library(MASS)
library(nnet)
data(housing)
house.mult <- multinom(Sat ~ Infl + Type + Cont, weights = Freq,
  data = housing)
relrelimp(house.mult, set1 = 2:3, set2 = 7, 
                      label1 = "Influence", label2 = "Contact",
                      response.cat1 = "Medium", response.cat2 = "High")
## Computes the relative contribution of Influence and Contact in 
## each of the two logistic regressions (Med/Low and High/Low), and
## compares those two relative-contribution measures.

Display a Data Frame in a Tk Text Widget

Description

Displays the contents of a data frame in a modeless Tk text window, for inspection. Objects not of class data.frame, for example objects of class table, or matrix, are coerced using as.data.frame prior to display.

Usage

showData(dataframe, 
       colname.bgcolor = "grey50", 
       rowname.bgcolor = "grey50", 
       body.bgcolor = "white", 
       colname.textcolor = "white", 
       rowname.textcolor = "white", 
       body.textcolor = "black",
       font = "Courier 12", 
       maxheight = 30, 
       maxwidth = 80, 
       title = NULL,
       rowname.bar = "left",
       colname.bar = "top",
       rownumbers = FALSE, 
       placement = "-20-40",
       suppress.X11.warnings = TRUE)
showData(dataframe, 
       colname.bgcolor = "grey50", 
       rowname.bgcolor = "grey50", 
       body.bgcolor = "white", 
       colname.textcolor = "white", 
       rowname.textcolor = "white", 
       body.textcolor = "black",
       font = "Courier 12", 
       maxheight = 30, 
       maxwidth = 80, 
       title = NULL,
       rowname.bar = "left",
       colname.bar = "top",
       rownumbers = FALSE, 
       placement = "-20-40",
       suppress.X11.warnings = TRUE)

Arguments

`dataframe`	A data frame, or an object to which `as.data.frame()` can be validly applied
`colname.bgcolor`	A background colour for the variable-names panel
`rowname.bgcolor`	A background colour for the row-names panel
`body.bgcolor`	A background colour for the data
`colname.textcolor`	A colour for the variable names
`rowname.textcolor`	A colour for the row names
`body.textcolor`	A colour for the data
`font`	The text font used – should be a monospaced font
`maxheight`	The maximum number of rows to display
`maxwidth`	The maximum width of display, in characters
`title`	A title for the window. Default is to use the name of the dataframe as given in the call to `showData()`
`rowname.bar`	position of sidebar for row names, `"left"` or `"right"`, or `c("left","right")`, or `NULL`
`colname.bar`	position of column names, `"top"` or `"bottom"`, or `c("top","bottom")`, or `NULL`
`rownumbers`	logical, whether row numbers should be displayed
`placement`	Position of the bottom right corner of the window
`suppress.X11.warnings`	logical, if `TRUE` then any X11 warnings are suppressed

Value

Invisibly returns the Tk window containing the displayed data frame.

Note

An error results if the printed representation of dataframe exceeds the maximum allowed width of 10000 characters; see options.

Text can be copied from the Tk window to the system clipboard, using <Control-C> or via a right-click pop-up menu.

On some systems the window may take a few seconds to appear if the data frame is very large.

Author(s)

David Firth, [email protected]; with Rcmdr-specific features contributed by John Fox

Examples

## This cannot be run by example() but should be OK when pasted
## into an interactive R session
## Not run: 
data(mtcars)
showData(mtcars)
## End(Not run)
## This cannot be run by example() but should be OK when pasted
## into an interactive R session
## Not run: 
data(mtcars)
showData(mtcars)
## End(Not run)

Convert a Tcl List to R Character Vector

Description

Converts a brace-delimited list from Tcl into a character vector

Usage

Tcl.to.R(tcl.list)
Tcl.to.R(tcl.list)

Arguments

tcl.list

a character string

Value

a character vector

Author(s)

David Firth, [email protected]

Examples

Tcl.to.R("{apple} {banana} {pear}")
Tcl.to.R("{apple} {banana} {pear}")

Package 'relimp'

Help Index

Pick Subsets from a Vector

Description

Usage

Arguments

Details

Value

Author(s)

Examples

Convert a Character Vector to Tcl Format

Description

Usage

Arguments

Value

Author(s)

See Also

Examples

Relative Importance of Predictors in a Regression Model

Description

Usage

Arguments

Details

Value

Author(s)

References

See Also

Examples

Comparison of Relative Importances in a Multinomial Logit Model

Description

Usage

Arguments

Details

Value

Author(s)

See Also

Examples

Display a Data Frame in a Tk Text Widget

Description

Usage

Arguments

Value

Note

Author(s)

Examples

Convert a Tcl List to R Character Vector

Description

Usage

Arguments

Value

Author(s)

See Also

Examples