The experimental GEE procedure fits generalized linear models for longitudinal data by using the generalized estimating equations (GEE) estimation method of Liang and Zeger (1986). The GEE method fits a marginal model to longitudinal data and is commonly used to analyze longitudinal data when the population-average effect is of interest. The GEE procedure’s syntax is similar to that of the GENMOD procedure. You specify the response variable and the explanatory variables in a MODEL statement, and you specify the correlation structure of multivariate responses in a REPEATED statement.
Because missing data are common in longitudinal studies and can lead to biased parameter estimates when missing responses depend on previous responses, PROC GEE also implements observation-specific and subject-specific weighted estimating equations. Both weighted estimators provide unbiased and consistent estimates when data are missing at random.
The ICPHREG procedure fits proportional hazards regression models to interval-censored data. You can fit models that have a variety of configurations with respect to the baseline hazard function, including the piecewise constant model and the cubic spline model. PROC ICPHREG maximizes the full likelihood instead of the Cox partial likelihood to estimate the regression coefficients. Standard errors of the estimates are obtained by inverting the observed information matrix that is derived from the full likelihood.
The ICPHREG procedure compares most closely to the PHREG and LIFEREG procedures. All three procedures can fit proportional hazards models. The distinction lies in the types of data that they are designed to handle and whether the baseline hazard function is parameterized. PROC PHREG deals exclusively with right-censored data, and it adopts a semiparametric approach by leaving the baseline hazard function unspecified. Both PROC LIFEREG and PROC ICPHREG can handle interval-censored data, which is a generalization of right-censored data. PROC LIFEREG focuses on parametric analysis by using accelerated failure time models, and it fits proportional hazards models only by assuming a Weibull baseline hazard function. PROC ICPHREG offers several options to parameterize the baseline hazard function. PROC ICPHREG generalizes PROC LIFEREG to fit more flexible parametric models and also generalizes PROC PHREG to handle interval-censored data.
The SPP procedure analyzes spatial point patterns. The broad goal of spatial point pattern analysis is to describe the occurrence of events (observations) that compose the pattern. The event locations are a discrete realization of a random spatial process. Therefore, the analysis goal is to investigate and characterize the original spatial process that generated the events in the spatial point pattern.
The SPP procedure enables you to specify the study area as a window, or you can rely on the input data coordinates to automatically compute a suitable study area by using the Ripley-Rasson window estimator. You can perform exploratory analysis of spatial point patterns by using the F, G, J, K, L, and PCF distance functions, which compare the empirical function distributions to the theoretical homogeneous Poisson process. PROC SPP enables you to perform nonparametric intensity estimation by using different types of kernels, and it supports adaptive kernel estimation. In addition, PROC SPP enables you to fit parametric inhomogeneous Poisson process models to perform model validation by using goodness-of-fit testing and a variety of residual diagnostics.