memisc (version 0.11-6)

mtable: Comparative Table of Model Estimates

Description

mtable produces a table of estimates for several models.

Usage

mtable(...,coef.style=getOption("coef.style"),
           summary.stats=TRUE,
           factor.style=getOption("factor.style"),
           getSummary=function(obj,...)UseMethod("getSummary"),
           float.style=getOption("float.style"),
           digits=min(3,getOption("digits")))
## S3 method for class 'mtable':
format(x,
          coef.title="Coefficients",
          summary.title="Summaries",
          colsep="\t",
          rowsep="\n",
          trim=TRUE,
          trimleft=trim,
          trimright=trim,
          center.at=NULL,
          align.integers=c("dot","right","left"),
          topsep="",
          bottomsep="",
          sectionsep="",
          forLaTeX=FALSE,
          useDcolumn=TRUE,
          colspec=if(useDcolumn) paste("D{.}{",LaTeXdec,"}{",ddigits,"}",sep="") else "r",
          LaTeXdec=".",
          ddigits="-1",
          useBooktabs=TRUE,
          toprule=if(useBooktabs) "\toprule" else "\hline\hline",
          midrule=if(useBooktabs) "\midrule" else "\hline",
          bottomrule=if(useBooktabs) "\bottomrule" else "\hline\hline",
          interaction.sep = if(forLaTeX) " $\times$ " else " x ",
          ...
          )
## S3 method for class 'mtable':
print(x,trim=FALSE,center.at=getOption("OutDec"),
      colsep=" ",
      topsep="=",bottomsep="=",sectionsep="-",...)
## S3 method for class 'mtable':
toLatex(object,...)
write.mtable(object,file="",...)
## S3 method for class 'mtable':
relabel(x,...)

Arguments

...

as argument to mtable: several model objects, e.g. of class lm; as argument to print.mtable, toLatex.mtable, write.mtable: further arguments passed t

coef.style

a character string which specifies the style of coefficient values, whether standard errors, Wald/t-statistics, or significance levels are reported, etc. See coef.style.

summary.stats

if FALSE, no summary statistics are repored. If TRUE, all summary statistics produced by getSummary are reported. This argument may also contain a character vector with the names of th

factor.style

a character string that specifies the style in which factor contrasts are labled. See factor.style.

getSummary

a function that computes model-related statistics that appear in the table. See getSummary.

float.style

default format for floating point numbers if no format is specified by coef.style; see {float.style}.

digits

number of significant digits if not specified by the template returned from getCoefTemplate getSummaryTemplate

x, object

an object of class mtable

coef.title

a character vector, the title for the reported coefficients.

summary.title

a character vector, the title for the reported model summaries.

colsep

a character string which seperates the columns in the output.

rowsep

a character string which seperates the rows in the output.

trim

should leading and trailing spaces be trimmed?

trimleft

should leading spaces be trimmed?

trimright

should trailing spaces be trimmed?

center.at

a character string on which resulting values are centered. Typically equal to ".". This is the default when forLaTeX==TRUE. If NULL, reported values are not centered.

align.integers

how to align integer values.

topsep

a character string that is recycled to a top rule.

bottomsep

a character string that is recycled to a bottom rule.

sectionsep

a character string that is recycled to seperate coefficients from summary statistics.

forLaTeX

should LaTeX code be produced?

useDcolumn

should the dcolumn LaTeX package be used?

colspec

LaTeX table column format specifyer(s).

LaTeXdec

the decimal point in the final LaTeX output.

ddigits

alignment specification or digits after the decimal point.

useBooktabs

should the booktabs LaTeX package be used?

toprule

appearance of the top border of the LaTeX tabular environment.

midrule

how are coefficients and summary statistics separated in the LaTeX tabular environment.

bottomrule

appearance of the bottom border of the LaTeX tabular environment.

interaction.sep

a character string that separates factors that are involved in an interaction effect

file

a file where to write to; defaults to console output.

Value

• A call to mtable results in an object that inherits from mtable with the following components:
• coefficients
  an array that contains the model coefficients.
• summaries
  a matrix that contains the model summaries.
• In addition the mtable object has attributes as.row, as.col, kill.col, kill.header, and coef.dim necessary for formatting.

Details

mtable constructs a table of estimates for regression-type models. format.mtable formats suitable for use with output or conversion functions such as print.mtable, toLatex.mtable, or write.mtable.

Examples

lm0 <- lm(sr ~ pop15 + pop75,              data = LifeCycleSavings)
lm1 <- lm(sr ~                 dpi + ddpi, data = LifeCycleSavings)
lm2 <- lm(sr ~ pop15 + pop75 + dpi + ddpi, data = LifeCycleSavings)

(mtable123 <- mtable("Model 1"=lm0,"Model 2"=lm1,"Model 3"=lm2))

#
# This should result in the following output:
#
# Calls:
# Model 1: lm(formula = sr ~ pop15 + pop75, data = LifeCycleSavings)
# Model 2: lm(formula = sr ~ dpi + ddpi, data = LifeCycleSavings)
# Model 3: lm(formula = sr ~ pop15 + pop75 + dpi + ddpi, data = LifeCycleSavings)
#
# ==================================================
#                  Model 1     Model 2     Model 3
# --------------------------------------------------
# Coefficients
# (Intercept)      30.628***    6.360***   28.566***
#                  (7.409)     (1.252)     (7.355)
# pop15            -0.471**                -0.461**
#                  (0.147)                 (0.145)
# pop75            -1.934                  -1.691
#                  (1.041)                 (1.084)
# dpi                           0.001      -0.000
#                              (0.001)     (0.001)
# ddpi                          0.529*      0.410*
#                              (0.210)     (0.196)
# --------------------------------------------------
# Summaries
# R-squared         0.262       0.162       0.338
# adj. R-squared    0.230       0.126       0.280
# sigma             3.931       4.189       3.803
# F                 8.3         4.5         5.8
# p                 0.001       0.016       0.001
# Log-likelihood -137.8      -141.0      -135.1
# Deviance        726.2       824.7       650.7
# AIC             283.7       290.0       282.2
# BIC             291.3       297.7       293.7
# N                50          50          50
# ==================================================
#
#

# Some other output formats

write.mtable(mtable123)
file123 <- "mtable123.txt"
write.mtable(mtable123,file=file123)
file.show(file123)
toLatex(mtable123)
texfile123 <- "mtable123.tex"
cat(toLatex(mtable123),sep="\n",file=texfile123)
file.show(texfile123)
#
# You should see:
# \%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%
# \%
# \% Calls:
# \% Model 1:   lm(formula = sr ~ pop15 + pop75, data = LifeCycleSavings)
# \% Model 2:   lm(formula = sr ~ dpi + ddpi, data = LifeCycleSavings)
# \% Model 3:   lm(formula = sr ~ pop15 + pop75 + dpi + ddpi, data = LifeCycleSavings)
# \%
# \%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%\%
# \begin{tabular}{lD{.}{.}{-1}D{.}{.}{-1}D{.}{.}{-1}}
# \toprule
#  & \multicolumn{1}{c}{Model 1}
#  & \multicolumn{1}{c}{Model 2}
#  & \multicolumn{1}{c}{Model 3}
# \\\\
# \midrule
# Coefficients    &                 &                 &                \\\\
# (Intercept)     &   30.628^{***}  &    6.360^{***}  &   28.566^{***} \\\\
#                 &   (7.409)       &   (1.252)       &   (7.355)      \\\\
# pop15           &   -0.471^{**}   &                 &   -0.461^{**}  \\\\
#                 &   (0.147)       &                 &   (0.145)      \\\\
# pop75           &   -1.934        &                 &   -1.691       \\\\
#                 &   (1.041)       &                 &   (1.084)      \\\\
# dpi             &                 &    0.001        &   -0.000       \\\\
#                 &                 &   (0.001)       &   (0.001)      \\\\
# ddpi            &                 &    0.529^{*}    &    0.410^{*}   \\\\
#                 &                 &   (0.210)       &   (0.196)      \\\\
# \midrule
# Summaries       &                 &                 &                \\\\
# R-squared       &    0.262        &    0.162        &    0.338       \\\\
# adj. R-squared  &    0.230        &    0.126        &    0.280       \\\\
# sigma           &    3.931        &    4.189        &    3.803       \\\\
# F               &    8.3          &    4.5          &    5.8         \\\\
# p               &    0.001        &    0.016        &    0.001       \\\\
# Log-likelihood  & -137.8          & -141.0          & -135.1         \\\\
# Deviance        &  726.2          &  824.7          &  650.7         \\\\
# AIC             &  283.7          &  290.0          &  282.2         \\\\
# BIC             &  291.3          &  297.7          &  293.7         \\\\
# N               &   50            &   50            &   50           \\\\
# \bottomrule
# \end{tabular}
#

berkeley <- aggregate(wtable(Admit,Freq)~.,data=UCBAdmissions)

berk0 <- glm(cbind(Admitted,Rejected)~1,data=berkeley,family="binomial")
berk1 <- glm(cbind(Admitted,Rejected)~Gender,data=berkeley,family="binomial")
berk2 <- glm(cbind(Admitted,Rejected)~Gender+Dept,data=berkeley,family="binomial")

mtable(berk0,berk1,berk2)

#
# This should result in the following output:
# 
# Calls:
# berk0: glm(formula = cbind(Admitted, Rejected) ~ 1, family = "binomial",
#     data = berkeley)
# berk1: glm(formula = cbind(Admitted, Rejected) ~ Gender, family = "binomial",
#     data = berkeley)
# berk2: glm(formula = cbind(Admitted, Rejected) ~ Gender + Dept, family = "binomial",
#     data = berkeley)
#
# =======================================================
#                        berk0       berk1       berk2
# -------------------------------------------------------
# Coefficients
# (Intercept)           -0.457***   -0.220***    0.582***
#                     (-14.972)    (-5.675)     (8.436)
# Gender: Female-Male               -0.610***    0.100
#                                  (-9.553)     (1.235)
# Dept: B-A                                     -0.043
#                                              (-0.395)
# Dept: C-A                                     -1.263***
#                                             (-11.841)
# Dept: D-A                                     -1.295***
#                                             (-12.234)
# Dept: E-A                                     -1.739***
#                                             (-13.792)
# Dept: F-A                                     -3.306***
#                                             (-19.452)
# -------------------------------------------------------
# Summaries
# McFadden R-sq.                     0.107       0.977
# Cox-Snell R-sq.                    0.020       0.172
# Nagelkerke R-sq.                   0.116       0.979
# phi                    1.000       1.000       1.000
# Likelihood-ratio                  93.4       856.9
# p                                  0.000       0.000
# Log-likelihood      -473.0      -426.3       -44.6
# Deviance             877.1       783.6        20.2
# AIC                  948.0       856.5       103.1
# BIC                  954.4       869.4       148.1
# N                   4526        4526        4526
# =======================================================

mtable(berk0,berk1,berk2,coef.style="stat")
mtable(berk0,berk1,berk2,coef.style="ci")
mtable(berk0,berk1,berk2,coef.style="ci.vertical")
mtable(berk0,berk1,berk2,coef.style="ci.horizontal")
mtable(berk0,berk1,berk2,coef.style="all")
mtable(berk0,berk1,berk2,coef.style="all.nostar")

mtable(by(berkeley,berkeley$Dept,function(x)glm(cbind(Admitted,Rejected)~Gender,data=x,family="binomial")))
#
# This should result in the following output:
#
# Calls:
# A: glm(formula = cbind(Admitted, Rejected) ~ Gender, family = "binomial",
#     data = x)
# B: glm(formula = cbind(Admitted, Rejected) ~ Gender, family = "binomial",
#     data = x)
# C: glm(formula = cbind(Admitted, Rejected) ~ Gender, family = "binomial",
#     data = x)
# D: glm(formula = cbind(Admitted, Rejected) ~ Gender, family = "binomial",
#     data = x)
# E: glm(formula = cbind(Admitted, Rejected) ~ Gender, family = "binomial",
#     data = x)
# F: glm(formula = cbind(Admitted, Rejected) ~ Gender, family = "binomial",
#     data = x)
# 
# =====================================================================================
#                         A          B          C          D          E          F
# -------------------------------------------------------------------------------------
# Coefficients
# (Intercept)           0.492***   0.534***  -0.536***  -0.704***  -0.957***  -2.770***
#                      (0.072)    (0.088)    (0.115)    (0.104)    (0.162)    (0.220)
# Gender: Female-Male   1.052***   0.220     -0.125      0.082     -0.200      0.189
#                      (0.263)    (0.438)    (0.144)    (0.150)    (0.200)    (0.305)
# -------------------------------------------------------------------------------------
# Summaries
# McFadden R-sq.        1.000      1.000      1.000      1.000      1.000      1.000
# Cox-Snell R-sq.       0.020      0.000      0.001      0.000      0.002      0.001
# Nagelkerke R-sq.      1.000      1.000      1.000      1.000      1.000      1.000
# phi                   1.000      1.000      1.000      1.000      1.000      1.000
# Likelihood-ratio     19.1        0.3        0.8        0.3        1.0        0.4
# p                     0.000      0.611      0.386      0.585      0.320      0.536
# Log-likelihood       -5.9       -5.1       -6.4       -6.3       -5.8       -4.9
# Deviance              0.0        0.0       -0.0        0.0        0.0       -0.0
# AIC                  15.7       14.3       16.9       16.6       15.6       13.8
# BIC                  25.4       23.0       26.5       26.0       24.3       23.0
# N                   933        585        918        792        584        714
# =====================================================================================



mtable(By(~Gender,glm(cbind(Admitted,Rejected)~Dept,family="binomial"),data=berkeley))
#
# This should result in the following output:
#
# Male: glm(formula = cbind(Admitted, Rejected) ~ Dept, family = "binomial")
# Female: glm(formula = cbind(Admitted, Rejected) ~ Dept, family = "binomial")
# 
# ========================================
#                     Male       Female
# ----------------------------------------
# Coefficients
# (Intercept)         0.492***    1.544***
#                    (0.072)     (0.253)
# Dept: B-A           0.042      -0.790
#                    (0.113)     (0.498)
# Dept: C-A          -1.028***   -2.205***
#                    (0.135)     (0.267)
# Dept: D-A          -1.196***   -2.166***
#                    (0.126)     (0.275)
# Dept: E-A          -1.449***   -2.701***
#                    (0.177)     (0.279)
# Dept: F-A          -3.262***   -4.125***
#                    (0.231)     (0.330)
# ----------------------------------------
# Summaries
# McFadden R-sq.      1.000       1.000
# Cox-Snell R-sq.     0.174       0.136
# Nagelkerke R-sq.    1.000       1.000
# phi                 1.000       1.000
# Likelihood-ratio  514.8       268.9
# p                   0.000       0.000
# Log-likelihood    -18.4       -16.1
# Deviance           -0.0         0.0
# AIC                48.7        44.2
# BIC                84.1        77.3
# N                2691        1835
# ========================================