Code: extended model support for mtable

March 22, 2011
By

(This article was first published on Left Censored » R, and kindly contributed to R-bloggers)

I finally got around to organizing and packaging my complete set of extended model support for mtable in Martin Elff’s memisc library. Here is a list of the models supported:

  • coxph, survreg – Cox proportional hazards models and parametric survival models from the survival library.
  • aftreg, phreg, weibreg – parametric AFT, proportional hazards, and Weibull models available from the eha library.
  • mer – multilevel mixed effects models as provided by the lme4 package. The model objects produced in the beta lme4a package are not supported.

Here’s the file: mtable-ext

It’s extremely easy to use, though I am sure there are some bugs and not all model features get reported. Here is an example of how to compare three Cox proportional hazard models as estimated with coxph (yes, I know, the Breslow and Exact models are the same; Exact was refusing to converge so I just stuck the Breslow model in to get the example done):

library(survival)
library(memisc)
source("mtable-ext.R")

> Efron <- coxph(Surv(start, end, switch) ~ Poll + Government + OldParty +
+                Experience + Right + Members, robust = TRUE, data = Sejm,
+                method = "efron")

> Breslow <- coxph(Surv(start, end, switch) ~ Poll + Government + OldParty +
+                  Experience + Right + Members, robust = TRUE, data = Sejm,
+                  method = "breslow")

> Exact <- coxph(Surv(start, end, switch) ~ Poll + Government + OldParty +
+                Experience + Right + Members, robust = TRUE, data = Sejm,
+                method = "exact")

mtable(Efron, Breslow, Exact)

[... some model call info ...]

=============================================
                  Efron    Breslow    Exact
---------------------------------------------
Poll            -2.120*** -1.931*** -1.931***
                (0.477)   (0.430)   (0.430)
Government      -1.603*   -1.461*   -1.461*
                (0.706)   (0.684)   (0.684)
OldParty        -0.186    -0.279    -0.279
                (0.243)   (0.209)   (0.209)
Experience      -0.195    -0.198    -0.198
                (0.167)   (0.156)   (0.156)
Right            1.427***  1.290***  1.290***
                (0.198)   (0.178)   (0.178)
Members         -0.002    -0.003    -0.003
                (0.007)   (0.007)   (0.007)
---------------------------------------------
Log-likelihood   -818.487  -832.625  -832.625
AIC              1648.973  1677.250  1677.250
BIC              1701.297  1729.574  1729.574
N               45280     45280     45280
=============================================

To convert the mtable object into a text{LaTeX{}} table, you just need to apply the toLatex function:

toLatex(mtable(Efron, Breslow, Exact))

Of course, many formatting options are available. See the documentation for details.

To leave a comment for the author, please follow the link and comment on his blog: Left Censored » R.

R-bloggers.com offers daily e-mail updates about R news and tutorials on topics such as: visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...



If you got this far, why not subscribe for updates from the site? Choose your flavor: e-mail, twitter, RSS, or facebook...

Tags: , ,

Comments are closed.