Building a Regression Model in R – Use #Rstats

One of the most commonly used uses of Statistical Software is building models, and that too logistic regression models for propensity in marketing of goods and services.

If building a model is what you do-here is a brief easy essay on how to build a model in R.

1) Packages to be used-

For smaller datasets

use these

CAR Package http://cran.r-project.org/web/packages/car/index.html
GVLMA Package http://cran.r-project.org/web/packages/gvlma/index.html
ROCR Package http://rocr.bioinf.mpi-sb.mpg.de/
Relaimpo Package
DAAG package
MASS package
Bootstrap package
Leaps package

Also see

http://cran.r-project.org/web/packages/rms/index.html or RMS package

rms works with almost any regression model, but it was especially written to work with binary or ordinal logistic regression, Cox regression, accelerated failure time models, ordinary linear models, the Buckley-James model, generalized least squares for serially or spatially correlated observations, generalized linear models, and quantile regression.

For bigger datasets also see Biglm http://cran.r-project.org/web/packages/biglm/index.html and RevoScaleR packages.

http://www.revolutionanalytics.com/products/enterprise-big-data.php

2) Syntax

outp=lm(y~x1+x2+xn,data=dataset) Model Eq
summary(outp) Model Summary
par(mfrow=c(2,2)) + plot(outp) Model Graphs
vif(outp) MultiCollinearity
gvlma(outp) Heteroscedasticity using GVLMA package
outlierTest (outp) for Outliers
predicted(outp) Scoring dataset with scores
anova(outp)
> predict(lm.result,data.frame(conc = newconc), level = 0.9, interval = “confidence”)