Normal-based confidence intervals for a parameter of interest are inaccurate when the sampling distribution of the estimate is nonnormal. The technique known as profile likelihood can produce confidence intervals with better coverage. It may be used when the model includes only the variable of interest or several other variables in addition. Profile-likelihood confidence intervals are particularly useful in nonlinear models. The command pllf computes and plots the maximum likelihood estimate and profile likelihood–based confidence interval for one parameter in a wide variety of regression models.
CoxD. R.1970. Analysis of Binary Data.London: Methuen.
2.
RoystonP.2001. Flexible parametric alternatives to the Cox model, and more. Stata Journal1: 1–28.
3.
RoystonP., and ParmarM. K. B.2002. Flexible parametric proportional-hazards and proportional-odds models for censored survival data, with application to prognostic modelling and estimation of treatment effects. Statistics in Medicine21: 2175–2197.
4.
SauerbreiW., and RoystonP.1999. Building multivariable prognostic and diagnostic models: Transformation of the predictors using fractional polynomials. Journal of the Royal Statistical Society, Series A162: 71–94.
5.
VenzonD. J., and MoolgavkarS. H.1988. A method for computing profile-likelihood-based confidence intervals. Applied Statistics37: 87–94.