We discuss methods for computing confidence intervals for predictions and discrete changes in predictions for regression models for categorical outcomes. The methods include endpoint transformation, the delta method, and bootstrapping. We also describe an update to prvalue and prgen from the SPost package, which adds the ability to compute confidence intervals. The article provides several examples that illustrate the application of these methods.
CoxC., and MaG.1995. Asymptotic confidence bands for generalized nonlinear re-gresion models. Biometrics51: 142–150.
2.
EfronB., and TibshiraniR.J.1986. Bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy. Statistical Science1: 54–75.
3.
EfronB.1993. An Introduction to the Bootstrap.New York: Chapman & Hall.
4.
GuanW.2003. From the help desk: bootstrapped standard errors. Stata Journal3: 71–80.
5.
LiaoT. F.2000. Estimated precision for predictions from generalized linear models in sociological research. Quality and Quantity34: 137–152.
6.
LongJ. S.1997. Regression Models for Categorical and Limited Dependent Variables.Thousand Oaks, CA: Sage.
7.
LongJ. S., and FreeseJ.2006. Regression Models for Categorical Dependent Variables Using Stata.2nd ed. College Station, TX: Stata Press.
8.
MooneyC. Z., and DuvalR.D.1993. Bootstrapping: A Nonparametric Approach to Statistical Inference.Newbury Park, CA: Sage.
9.
PoiB.2004. From the help desk: some bootstrapping techniques. Stata Journal4: 312–328.
