Interactive Interpretation of Linear Regression Models

Description

The implementation is based on the package rpanel and provides an interactive visualization of effects, response functions and marginal effects for different kinds of regression models. Major features are the interactive approach and the handling of the effects of categorical covariates: if two or more factors are used as covariates every combination of the levels of each factor (referred to as groups) is treated separately. The automatic calculation of marginal effects and a number of possibilities to customize the graphical output are useful features as well. In the accompanying vignette the usage is described in detail. Please refer to the documentation of the generic function fxInteractive for available methods.

Details

Author(s)

Martin Meermeyer m.meermeyer@gmail.com

References

Bowman, A., Crawford, E., Alexander, G., and Bowman, R. (2007). rpanel: Simple interactive controls for R functions using the tcltk package. Journal of Statistical Software, 17(9), 1-18.

See Also

The noninteractive visualization of the results for various types of regression models can be achieved with the package effects.

Examples

## Not run: 
### Metric dependent variable ###
data("munichrent03")
model.rent <- lm(rent ~ yearc + I(yearc^2) + rooms + area*location + upkitchen,
    data=munichrent03)
fxInteractive(model.rent)

### Binary dependent variable ###
data("creditdata")
model.cd <- glm(credit ~ amount + I(amount^2) + age + duration*teleph + housing,
    family=binomial, data=creditdata)
fxInteractive(model.cd)

## End(Not run)

Credit Default Data

Description

Stratified sample of 1,000 credit receivers (in the following cr) from a private bank in southern Germany. The data were sampled at the end of the 1970s, therefore the credit amount is given in DEM (1.95583 DEM = 1 EUR).

Usage

data("creditdata")

Format

A data frame with 1,000 observations on the following 21 variables.

credit

Factor with levels "good" (no problems occur in redemption) and "bad" (problems occur in redemption).

amount

Credit amount in DEM (numeric).

duration

Duration of lent term in months (numeric).

age

Age of the cr in years (numeric).

addcredit

Has the cr additional credits at other vendors? Ordered factor with levels "none" (no additional credits), "bank" (at other banks) and "bargain" (at other vendors than banks).

asset

Most valuable asset of the cr. Ordered factor with levels "none" (no valuable assets), "car" (car), "savings" (savings) and "prop" (freehold property).

bail

Bailsman available? Ordered factor with levels "none" (no bailsmen), "applicant" (bailsman is additional applicant for the credit) and "yes" (external bailsman).

curacc

Current account of the cr. Ordered factor with levels "none" (no account), "debit" (account with dedit balance), "credit" (account with credit balance) and "wage" (account with periodical income).

empldur

Duration of employment of the cr at the current employer. Ordered factor with levels "none" (unemployed), "max1y" (less than 1 year), "max4y" (between 1 and 4 years), "max7y" (between 4 and 7 years) and "min7y" (more than 7 years).

housdur

For which period does the cr lives in current housing? Ordered factor with levels "max1y" (less than 1 year), "max4y" (between 1 and 4 years), "max7y" (between 4 and 7 years) and "min7y" (more than 7 years).

housing

Accommodation of the cr. Ordered factor with levels "social" (social accommodation), "rent" (rented accommodation) and "freehold" (freehold accommodation).

immigrant

Factor: Is the cr an immigrant?

installment

Monthly installment in % of disposable income. Ordered factor with levels "max20" (installment is less than 20% of disposable income), "max25" (installment is more than 20% and less than 25% of disposable income), "max35" (installment is more than 25% and less than 35% of disposable income) and "min35" (installment is more than 35% of disposable income).

intuse

Intended usage for the credit. Factor with levels "car-new" (buy a new car), "car-old" (buy an used car), "fittings" (buy fittings), "hifi" (buy radio or TV), "holiday" (pay a holiday), "househ" (buy new household items), "misc" (miscellaneous activities), "qualif" (pay a new qualification), "repair" (repair something) and "retrain" (pay a retrain).

job

Employment status of the cr. Ordered factor with levels "none" (unemployed), "unskilled" (unskilled worker), "medium" (intermediate position) and "higher" (leading position).

martsex

Martial status and sex of the cr. Factor with levels "f-div-m-sing" (female-divorced and male-single), "f-sing" (female-single), "m-div" (male-divorced) and "m-mar" (male-married).

moral

Payment moral of the cr. Factor with levels "alright" (no problems with credits in the past at the bank), "critical" (further credits at other banks or problems with account management), "none" (no credits previously or fully paid back), "yetalright" (no problems with current credits at the bank so far) and "problems" (problems with credits in the past).

pers

Number of persons who are entitled to maintenance from cr. Ordered factor with levels "max2" (up to 2) and "min3" (more than 2).

prevcred

Number of previous credits of the cr at the bank (including the actual). Ordered factor with levels "max1" (1 credit), "max3" (between 2 and 3), "max5" (between 4 and 5) and "min6" (6 or more).

savacc

Savings account of the cr. Ordered factor with levels "none" (no savings account), "max1h" (savings account with a balance up to 100 DEM), "max5h" (savings account with a balance between 100 DEM and 500 DEM), "max1t" (savings account with a balance between 500 DEM and 1.000 DEM) and "min1t" (savings account with a balance greater than 1,000 DEM).

teleph

Factor: Has the cr a landline?

Source

doi:10.5282/ubm/data.23

References

Fahrmeir, L., Hamerle, A., Tutz, G. (1996): Multivariate statistische Verfahren, 2nd edition. Berlin: de Gruyter. 390 ff.

Fahrmeir, L., Kneib, T., Lang, S., Marx, B. (2013): Regression: Models, Methods and Applications. Berlin: Springer.

Open Data LMU (2010): Kreditscoring zur Klassifikation von Kreditnehmern.

Examples

data("creditdata")
print(summary(creditdata))

par(ask=TRUE)
plot(credit ~ amount, data=creditdata)
plot(credit ~ duration, data=creditdata)
plot(credit ~ age, data=creditdata)
plot(credit ~ housing, data=creditdata)
plot(credit ~ intuse, data=creditdata)
plot(credit ~ teleph, data=creditdata)
par(ask=FALSE)

Combine Levels of Different Factors to Groups

Description

All levels of the factors from a data.frame provided are combined, each combination of factor levels is referred to as group.

Usage

factorCombinations(X, factor.sep = "|", level.sep = ".", count=TRUE)

Arguments

Value

A list with the following components:

Note

Function is mainly for internal use in LinRegInteractive but may be useful for other purposes as well.

Examples

# multiple factors
data("CO2")
print(factorCombinations(CO2))

# single factor
data("chickwts")
print(factorCombinations(chickwts))

Interactive Interpretation of Linear Regression Models

Description

fxInteractive is a generic function for the interactive visualization of the results of various types of regression models. Originally the function was developed to translate proposals for the interpretation of models for binary outcomes made by Hoetker (2007) into practice. The function invokes particular methods for different kinds of fitted-model objects, see the list of available methods in the details. The implementation is based on the package rpanel. Major features are the interactive approach and the handling of the effects of categorical covariates: if two or more factors are used as covariates every combination of the levels of each factor (referred to as groups) is treated separately. The automatic calculation of marginal effects and a number of possibilities to customize the graphical output are useful features as well. In the accompanying vignette the usage is described in detail.

Usage

fxInteractive(model, ...)

Arguments

Details

The only mandatory argument of the function is an appropriate fitted-model object. In this package version methods for the following regression models are available:

• Generalized linear models fitted with glm. The glm-method also works for generalized additive models fitted with gam (package gam and mgcv).

• Multiple linear regression models fitted with lm.

• Linear mixed-effects models fitted with lme (package nlme).

For the fitted-model objects the following prerequisites must be met:

• The model must contain at least one metric covariate.

• The model must be specified with the formula interface and the data frame containing the variables must be passed with the data argument.

• The categorical variables must be factors (ordered or unordered).

The basic idea of the function is the interactive usage. It is nevertheless easy to reproduce the plots for publication purposes, refer to section 5 of the vignette for this.

Users of the IDE RStudio may need to change the graphic device with options(device = "x11") before calling the function because in current versions of RStudio multiple graphic devices occasionally do not work.

Value

The default method just returns a notice. For supported fitted-model objects no object is returned. By calling the function usually a menu appears from which a metric covariate employed in the model must be selected. After choosing the covariate a graphic device which contains a termplot of the selected metric covariate and a GUI-panel to manipulate the plot will be opened. The GUI-panel has the following elements:

• A slider for each metric covariate.

• A radiobox to select the type of the current display (effect, response (if appropriate) and marginal effect).

• A checkbox to select the factor combinations (groups) to be displayed.

• A button to print tables of output to the console, see below.

The appearance of the panel can be controlled by a number of arguments, see section 6.4 of the vignette for details.

When the button is clicked four tables are printed to the console:

1. Table of coefficients obtained by the summary-method.

2. Table of the chosen values of the metric covariates and their ECDF-values in the dataset.

3. Table of the link and response function at the chosen values of the metric covariates for each group.

4. Table of marginal effects for each metric covariate at the chosen values of the metric covariates for each group.

By setting the argument latex2console to TRUE (see the documentation of the methods) the latter three tables are printed as LaTeX-code using functions provided by the package xtable (Dahl, 2012), see section 4 of the vignette for details.

The format of the text output and the layout of the plots can be controlled to a large extent, see section 6.1, 6.2 and 6.3 of the vignette for details.

The code to save plots in a platform independent way is adopted from the program archive accompanying Kruschke (2014) (URL: http://rpubs.com/NMA/9704).

Author(s)

Martin Meermeyer m.meermeyer@gmail.com

References

Bowman, A., Crawford, E., Alexander, G., and Bowman, R. (2007). rpanel: Simple interactive controls for R functions using the tcltk package. Journal of Statistical Software, 17(9), 1-18.

Dahl, D. B. (2014). xtable: Export tables to LaTeX or HTML. R package version 1.7-4.

Hoetker, G. (2007). The use of logit and probit models in strategic management research: Critical issues. Strategic Management Journal, 28(4), 331-343.

Kruschke, J. K. (2014). Doing Bayesian Data Analysis: A Tutorial with R, JAGS, and Stan, 2nd edition. Waltham (MA): Elsevier (Academic Press).

See Also

fxInteractive.glm is the corresponding method for generalized linear models fitted with glm.

fxInteractive.lm is the corresponding method for multiple linear regression models fitted with lm.

fxInteractive.lme is the corresponding method for linear mixed-effects models fitted with lme (package nlme).

The noninteractive visualization of the results for various types of regression models can be achieved with the package effects.

Interactive Interpretation of Generalized Linear Models

Description

This method implements proposals for the interpretation of models for binary outcomes made by Hoetker (2007) but works for other types of generalized linear models as well. The method is also suitable for generalized additive models fitted with gam (package gam and mgcv).

Usage

## S3 method for class 'glm'
fxInteractive(model, initial.values = as.list(NULL),
    preselect.var = NA, preselect.type = "link", preselect.groups = NULL,
    dev.height = 18, dev.width = 18, dev.width.legend = 8, dev.pointsize = 10,
    dev.defined = FALSE, ylim = NA, col = NA, lty = 1, lwd = 1, 
    main = NA, main.line = 1.5, xlab = NA, ylab = NA, 
    legend.add = TRUE, legend.space = legend.add, legend.only = FALSE,
    legend.pos = "center", legend.cex = 1, legend.width.factor = 1, 
    rug.ticksize = 0.02, rug.col = "black", vline.actual = TRUE, 
    pos.hlines = c(0, 0.5, 0), n.effects = 100, 
    autosave.plot = FALSE, snapshot.plot = FALSE, 
    graphics.filename = "LinRegIntPlot", graphics.numbering = !autosave.plot,
    graphics.type = "pdf", factor.sep = "|", level.sep = ".", 
    latex2console = FALSE, xtable.big.mark = ".", xtable.decimal.mark = ",", 
    xtable.digits = NULL, xtable.display = NULL, xtable.booktabs = FALSE,                 
    panel.title = "Generalized Linear Model", label.button = "Snapshot",
    label.slider.act = "Variable displayed: ", label.box.type = "Type",
    label.types = c("linear predictor", "response", "marginal effect"),
    label.box.groups = "Groups", 
    slider.width = 200, slider.height = 60, button.height = 30,
    box.type.height = 100, box.group.character.width = 7, 
    box.group.line.height = 28, dist.obj.width = 20,
    dist.obj.height = 10, ...)

Arguments

Details

The only mandatory argument of the function is a fitted-model object of class glm. For this object the following prerequisites must be met:

• The model must contain at least one metric covariate.

• The model must be specified with the formula interface and the data frame containing the variables must be passed with the data argument.

• The categorical variables must be factors (ordered or unordered).

Please refer to the documentation of fxInteractive or the vignette for more details.

Value

No object is returned, please refer to the documentation of fxInteractive or the vignette for more details.

Author(s)

Martin Meermeyer m.meermeyer@gmail.com

References

Hoetker, G. (2007). The use of logit and probit models in strategic management research: Critical issues. Strategic Management Journal, 28(4), 331-343.

Kleiber, C., Zeileis, A. (2008). Applied Econometrics with R. New York: Springer.

Examples

### Model specification ###
data("creditdata")

## Treat ordered factors as unordered factors.
options(contrasts=c("contr.treatment","contr.treatment"))

model.cd <- glm(credit ~ amount + I(amount^2) + age + duration*teleph + housing,
    family=binomial, data=creditdata)

## Not run: 

### Basic usage ###

## RStudio users may need to change the graphic device, see details.
options(device = "x11")

## Using defaults
fxInteractive(model.cd)

## Switch text output to LaTeX
fxInteractive(model.cd, latex2console = TRUE)

## Continental European number format in LaTeX output
fxInteractive(model.cd, latex2console = TRUE, decimal.mark = ",", big.mark = ".")

## Save plot as PDF to current working directory when 'Snapshot' is clicked
fxInteractive(model.cd, snapshot.plot = TRUE)

## Change color scheme and line types
fxInteractive(model.cd, col = rep(c(2,4),each = 3), lty = c(1,3,5))

## Change separation characters
fxInteractive(model.cd, factor.sep = "-", level.sep = ">")

## Suppress legend
fxInteractive(model.cd, legend.add = FALSE)

## Suppress rug plot
fxInteractive(model.cd, rug.ticksize = NA)

## Set initial values of metric covatiates 
fxInteractive(model.cd, initial.values = list(amount=5000, age=30, duration=24))

## Preselect covariate, plot type and groups
fxInteractive(model.cd, preselect.var = "duration", preselect.type = "response",
    preselect.groups = c(2,5))

## Preselect covariate and plot type and change axis annotations
fxInteractive(model.cd, preselect.var = "duration", preselect.type = "response", 
    xlab = "duration (months)", ylab = "probability of credit default")


### Visualization of statistical concepts ###

## Nonparametric effect 
require("splines")
model.cd.bs <- glm(credit ~ bs(amount) + age + duration*teleph + housing,
    family = binomial, data = creditdata)
fxInteractive(model.cd.bs, preselect.var = "amount")

## Generalized additive model
require("mgcv")
model.cd.mgcv <- gam(credit ~ s(amount) + age + duration*teleph + housing,
family = binomial, data = creditdata)
fxInteractive(model.cd.mgcv)

## Interaction effect (directly)
fxInteractive(model.cd, preselect.var = "duration")

## Interaction effect (indirectly)
fxInteractive(model.cd, preselect.var = "age")
# manipulate slider for 'duration'

## Quasi-complete separation
# example from Kleiber, Zeileis (2008), p. 130ff 
require("AER")
data("MurderRates")
model.mr <- glm(I(executions > 0) ~ time + income + noncauc + lfp + southern,
    family = binomial, data = MurderRates)
fxInteractive(model.mr, preselect.var = "income", preselect.type = "response")


### Additional examples ###

## Customize device for printing
fxInteractive(model.cd, 
    dev.width = 6,
    dev.width.legend = 4,
    dev.height = 6,
    dev.pointsize = 6,
    col = c("darkred","red","salmon","darkblue","blue","lightblue"),
    legend.width.factor = 1.1,
    vline.actual = FALSE,
    snapshot.plot = TRUE,
    graphics.filename = "creddefault-termplot",
    mar = c(2.5,2.5,1,1)+0.1,
    mgp = c(1.5,0.5,0),
    tcl = -0.3)

## Save predefined plot automatically
fxInteractive(model.cd,
    initial.values = list(amount=5000, duration=24, age=30),
    preselect.var = "duration",
    preselect.type = "link",
    autosave.plot = TRUE,
    graphics.filename = "fig-creddefault-duration-link",
    legend.width.factor = 1.05)

## Modifications for models with many groups
# Increase space for legend and squeeze panel controls
model.cd.moregroups <- glm(credit ~ amount + I(amount^2) + age 
    + duration*teleph + housing + job, family = binomial, data = creditdata)
fxInteractive(model.cd.moregroups, 
    dev.width.legend = 9,
    legend.cex = 1,
    box.type.height = 90,
    box.group.character.width = 6,
    box.group.line.height = 25,
    dist.obj.height = 2)

# Squeeze legend and  panel controls
model.cd.manygroups <- glm(credit ~ amount + I(amount^2) + age 
    + duration*teleph + housing + intuse, family = binomial, data = creditdata)
fxInteractive(model.cd.manygroups, 
    dev.width.legend = 6,
    legend.cex = 0.54,
    box.type.height = 90,
    box.group.character.width = 6,
    box.group.line.height = 25,
    dist.obj.height = 2)
# Note that checkbox for groups grows beyond screen

# Preselect groups and specifiy color and line types directly
model.cd.manygroups <- glm(credit ~ amount + I(amount^2) + age 
    + duration*teleph + housing + intuse, family = binomial, data = creditdata)
index.groups <- c(1,11,21,31,41,51)
vec.col <- NULL
vec.col[index.groups] <- c(1:6)
vec.lty <- NULL
vec.lty[index.groups] <- rep(c(1,2), each = 3)
fxInteractive(model.cd.manygroups,
preselect.var = "amount",
preselect.groups = index.groups,
col = vec.col,
lty = vec.lty)

## End(Not run)

Interactive Interpretation of Multiple Linear Regression Models

Description

This method facilitates the interpretation of multiple linear regression models and is a byproduct of the method fxInteractive.glm for generalized linear models.

Usage

## S3 method for class 'lm'
 fxInteractive(model, initial.values = as.list(NULL),
    preselect.var = NA, preselect.type = "effect", preselect.groups = NULL,
    dev.height = 18, dev.width = 18, dev.width.legend = 8, dev.pointsize = 10,
    dev.defined = FALSE, ylim = NA, col = NA, lty = 1, lwd = 1, 
    main = NA, main.line = 1.5, xlab = NA, ylab = NA, 
    legend.add = TRUE, legend.space = legend.add, legend.only = FALSE,
    legend.pos = "center", legend.cex = 1, legend.width.factor = 1, 
    rug.ticksize = 0.02, rug.col = "black", vline.actual = TRUE, 
    pos.hlines = c(0, 0), n.effects = 100, 
    autosave.plot = FALSE, snapshot.plot = FALSE, 
    graphics.filename = "LinRegIntPlot", graphics.numbering = !autosave.plot,
    graphics.type = "pdf", factor.sep = "|", level.sep = ".", 
    latex2console = FALSE, xtable.big.mark = ".", xtable.decimal.mark = ",", 
    xtable.digits = NULL, xtable.display = NULL, xtable.booktabs = FALSE,                 
    panel.title = "Linear Model", label.button = "Snapshot",
    label.slider.act = "Variable displayed: ", label.box.type = "Type",
    label.types = c("effect", "marginal effect"),
    label.box.groups = "Groups", 
    slider.width = 200, slider.height = 60, button.height = 30,
    box.type.height = 75, box.group.character.width = 7, 
    box.group.line.height = 28, dist.obj.width = 20,
    dist.obj.height = 10, ...)

Arguments

Details

The only mandatory argument of the function is a fitted-model object of class lm. For this object the following prerequisites must be met:

• The model must contain at least one metric covariate.

• The model must be specified with the formula interface and the data frame containing the variables must be passed with the data argument.

• The categorical variables must be factors (ordered or unordered).

Please refer to the documentation of fxInteractive or the vignette for more details.

Value

No object is returned, please refer to the documentation of fxInteractive or the vignette for more details.

Author(s)

Martin Meermeyer m.meermeyer@gmail.com

Examples

### Model specification ###
data("munichrent03")

## Treat ordered factors as unordered factors.
options(contrasts=c("contr.treatment","contr.treatment"))

model.rent <- lm(rent ~ yearc + I(yearc^2) + rooms + area*location + upkitchen,
    data=munichrent03)

## Not run: 

### Basic usage ###

## RStudio users may need to change the graphic device, see details.
options(device = "x11")

## Using defaults
fxInteractive(model.rent)

## Switch text output to LaTeX
fxInteractive(model.rent, latex2console = TRUE)

## Continental European number format in LaTeX output
fxInteractive(model.rent, latex2console = TRUE, xtable.decimal.mark = ",", xtable.big.mark = ".")

## Save plot as PDF to current working directory when 'Snapshot' is clicked
fxInteractive(model.rent, snapshot.plot = TRUE)

## Change color scheme and line types
fxInteractive(model.rent, col = rep(c(1,2,4),each = 2), lty = c(1,5))

## Change separation characters
fxInteractive(model.rent, factor.sep = "-", level.sep = ">")

## Suppress legend
fxInteractive(model.rent, legend.add = FALSE)

## Suppress rug plot
fxInteractive(model.rent, rug.ticksize = NA)

## Set initial values of metric covatiates 
fxInteractive(model.rent, initial.values = list(yearc=1990, rooms=3, area=80))

## Preselect covariate, plot type and groups
fxInteractive(model.rent, preselect.var = "yearc", preselect.type = "marginal",
    preselect.groups = c(2,4,6))

## Preselect covariate and plot type and change axis annotations
fxInteractive(model.rent, preselect.var = "yearc", preselect.type = "effect", 
    xlab = "year of construction", ylab = "net rent (EUR)")


### Visualization of statistical concepts ###

## Nonlinear effect 
fxInteractive(model.rent, preselect.var = "yearc")

## Nonparametric effect 
require("splines")
model.rent.bs <- lm(rent ~ bs(yearc) + rooms + area*location + upkitchen,
    data = munichrent03)
model.rent.bs$data <- munichrent03
fxInteractive(model.rent.bs, preselect.var = "yearc")
# Note that the data must be attached to the fitted-model object
# afterwards, see section 7 of the vignette for details.

## Interaction effect (directly)
fxInteractive(model.rent, preselect.var = "area")

## Interaction effect (indirectly)
fxInteractive(model.rent, preselect.var = "yearc")
# manipulate slider for 'area'


### Additional examples ###

## Customize device for printing
fxInteractive(model.rent, 
    dev.width = 6,
    dev.width.legend = 4,
    dev.height = 6,
    dev.pointsize = 6,
    col = c("red","darkred","blue","darkblue","green","darkgreen"),
    legend.width.factor = 1.1,
    vline.actual = FALSE,
    snapshot.plot = TRUE,
    graphics.filename = "munichrent-termplot",
    mar = c(2.5,2.5,1,1)+0.1,
    mgp = c(1.5,0.5,0),
    tcl =  -0.3)

## Save predefined plot automatically
fxInteractive(model.rent,
    initial.values = list(yearc=1990, rooms=3, area=80), 
    preselect.var = "area",
    preselect.type = "marginal",
    autosave.plot = TRUE,
    graphics.filename = "fig-rent-area-marg",
    legend.width.factor = 1.05)

## Modifications for models with many groups
# Increase space for legend and squeeze panel controls
model.rent.moregroups <- lm(rent ~ yearc + I(yearc^2) + rooms + area*location 
    + upkitchen + bathtile, data = munichrent03)
fxInteractive(model.rent.moregroups, 
    dev.width.legend = 9,
    legend.cex = 1,
    box.type.height = 65,
    box.group.character.width = 6,
    box.group.line.height = 28,
    dist.obj.height = 2)

# Squeeze legend and  panel controls
model.rent.manygroups <- lm(rent ~ yearc + I(yearc^2) + rooms + area 
    + district + upkitchen, data = munichrent03)
fxInteractive(model.rent.manygroups, 
    dev.width.legend = 6,
    legend.cex = 0.65,
    box.type.height = 90,
    box.group.character.width = 6,
    box.group.line.height = 25,
    dist.obj.height=2)
# Note that checkbox for groups grows beyond screen

# Preselect groups and specifiy color and line types directly
model.rent.manygroups <- lm(rent ~ yearc + I(yearc^2) + rooms + area 
    + district + upkitchen, data = munichrent03)
index.groups <- c(3,4,17,18,49,50)
vec.col <- NULL
vec.col[index.groups] <- c(1:6)
vec.lty <- NULL
vec.lty[index.groups] <- rep(c(1,2), each = 3)
fxInteractive(model.rent.manygroups,
preselect.var = "area",
preselect.groups = index.groups,
col = vec.col,
lty = vec.lty)

## End(Not run)

Interactive Interpretation of Linear Mixed-Effects Models

Description

This method facilitates the interpretation of linear mixed-effects models and is a byproduct of the method fxInteractive.glm for generalized linear models.

Usage

## S3 method for class 'lme'
 fxInteractive(model, predict.lme.level = 0, initial.values = as.list(NULL),
    preselect.var = NA, preselect.type = "effect", preselect.groups = NULL,
    dev.height = 18, dev.width = 18, dev.width.legend = 8, dev.pointsize = 10,
    dev.defined = FALSE, ylim = NA, col = NA, lty = 1, lwd = 1, 
    main = NA, main.line = 1.5, xlab = NA, ylab = NA, 
    legend.add = TRUE, legend.space = legend.add, legend.only = FALSE,
    legend.pos = "center", legend.cex = 1, legend.width.factor = 1, 
    rug.ticksize = 0.02, rug.col = "black", vline.actual = TRUE, 
    pos.hlines = c(0, 0), n.effects = 100, 
    autosave.plot = FALSE, snapshot.plot = FALSE, 
    graphics.filename = "LinRegIntPlot", graphics.numbering = !autosave.plot,
    graphics.type = "pdf", factor.sep = "|", level.sep = ".", 
    latex2console = FALSE, xtable.big.mark = ".", xtable.decimal.mark = ",", 
    xtable.digits = NULL, xtable.display = NULL, xtable.booktabs = FALSE,                 
    panel.title = "Linear Mixed-Effects Model", label.button = "Snapshot",
    label.slider.act = "Variable displayed: ", label.box.type = "Type",
    label.types = c("effect", "marginal effect"),
    label.box.groups = "Groups", 
    slider.width = 200, slider.height = 60, button.height = 30,
    box.type.height = 75, box.group.character.width = 7, 
    box.group.line.height = 28, dist.obj.width = 20,
    dist.obj.height = 10, ...)

Arguments

Details

The only mandatory argument of the function is a fitted-model object of class lme (package nlme). For this object the following prerequisites must be met:

• The model must contain at least one metric covariate.

• The model must be specified with the formula interface and the data frame containing the variables must be passed with the data argument.

• The categorical variables must be factors (ordered or unordered).

In this method the additional argument predict.lme.level is responsible for the level of grouping to be displayed, see the documentation of predict.lme.

Please refer to the documentation of fxInteractive or the vignette for more details.

Value

No object is returned, please refer to the documentation of fxInteractive or the vignette for more details.

Author(s)

Martin Meermeyer m.meermeyer@gmail.com

See Also

The examples in fxInteractive.glm and fxInteractive.lm are more elaborated and show different aspects of usage and customization which are valid for this method, too.

Examples

require("nlme")
data("Orthodont")
LMEmod <- lme(distance ~ age + Sex, data = Orthodont, random = ~ age | Subject)

## Not run: 
## RStudio users may need to change the graphic device, see details.
options(device = "x11")

## Grouping level 0 is used by default.
fxInteractive(LMEmod, legend.cex=0.6)

## Display grouping level 1
fxInteractive(LMEmod, predict.lme.level=1, legend.cex=0.6)

## For individual specific factors only observable factor combinations are meaningful 
# the grouping variable must be the last column
combs.count    <- factorCombinations(Orthodont[,c(4,3)])$counts 
combs.observed <- combs.count > 0
fxInteractive(LMEmod, predict.lme.level=1, preselect.groups=combs.observed)

## End(Not run)

Data for the Rent Index 2003 in Munich, Germany

Description

Sample of 2,053 appartments from the data collected for the preparation of the Munich rent index 2003.

Usage

data("munichrent03")

Format

A data frame with 2,053 observations on the following 12 variables.

rent

Net rent in EUR (numeric).

rentsqm

Net rent per square meter in EUR (numeric).

area

Floor area in square meters (numeric).

rooms

Number of rooms (numeric).

yearc

Year of construction (numeric).

bathextra

Factor: High quality equipment in the bathroom?

bathtile

Factor: Bathroom tiled?

cheating

Factor: Central heating available?

district

Urban district where the apartment is located. Factor with 25 levels: "All-Umenz" (Allach - Untermenzing), "Alt-Le" (Altstadt - Lehel), "Au-Haid" (Au - Haidhausen), "Au-Lo-La" (Aubing - Lochhausen - Langwied), "BamLaim" (Berg am Laim), "Bogenh" (Bogenhausen), "Feld-Has" (Feldmoching - Hasenbergl), "Had" (Hadern), "Laim" (Laim), "Lud-Isar"(Ludwigsvorstadt - Isarvorstadt), "Maxvor" (Maxvorstadt), "Mil-AmH" (Milbertshofen - Am Hart), "Moos" (Moosach), "Neuh-Nymp" (Neuhausen - Nymphenburg), "Obgies" (Obergiesing), "Pas-Obmenz" (Pasing - Obermenzing), "Ram-Per" (Ramersdorf - Perlach), "SchwWest" (Schwabing West), "Schwab-Frei" (Schwabing - Freimann), "Schwanth" (Schwanthalerhoehe), "Send" (Sendling), "Send-West" (Sendling - Westpark), "Th-Ob-Fo-Fu-So" (Thalkirchen - Obersendling - Forstenried - Fuerstenried - Solln), "Trud-Riem" (Trudering - Riem) and "Ugies-Har" (Untergiesing - Harlaching).

location

Quality of location. Ordered factor with levels "normal", "good" and "top".

upkitchen

Factor: Upscale equipment in kitchen?

wwater

Factor: Hot water supply available?

Source

doi:10.5282/ubm/data.2

References

Fahrmeir, L., Kneib, T., Lang, S., Marx, B. (2013): Regression: Models, Methods and Applications. Berlin: Springer.

Fahrmeir, L., Kuenstler, R., Pigeot, I., Tutz, G. (2004): Statistik: der Weg zur Datenanalyse, 5th edition. Berlin: Springer.

Open Data LMU (2003): Muenchner Mietspiegel 2003.

Examples

data("munichrent03")
print(summary(munichrent03))

par(ask=TRUE)
plot(munichrent03[,1:5])
plot(rentsqm ~ bathextra, data=munichrent03)
plot(rentsqm ~ bathtile, data=munichrent03)
plot(rentsqm ~ cheating, data=munichrent03)
plot(rentsqm ~ location, data=munichrent03)
plot(rentsqm ~ upkitchen, data=munichrent03)
plot(rentsqm ~ wwater, data=munichrent03)

oldpar <- par(no.readonly = TRUE)
par(las=3, mar=c(8,4,4,2)+0.1)
plot(rentsqm ~ district, xlab=NA, data=munichrent03)
par(oldpar)

par(ask=FALSE)

Wrapper Function for rp.slider

Description

Wrapper function to provide a variable name of type character to rp.slider.

Usage

wrapRpSlider(panel, variable, from, to, action, title=NA, 
log = FALSE, showvalue = FALSE, resolution = 0, 
initval = NULL, pos = NULL, horizontal = TRUE)

Arguments

See documentation of rp.slider. The argument variable must be of type character here.

Value

A function call of rp.slider which can be evaluated.

Note

Function is only for internal use in LinRegInteractive.