SAS/Ets User's Guide 56. Provides detailed reference material for using SAS/ETS software and guides you through the analysis and forecasting of features such as univariate and multivariate time series, cross-sectional time series, seasonal adjustments, multiequational nonlinear models, discrete choice models, limited dependent variable models, portfolio analysis, and generation of financial reports, with introductory and advanced examples for each procedure. You can also find complete information about two easy-to-use point-and-click applications: the Time Series Forecasting System, for automatic and interactive time series modeling and forecasting, and the Investment Analysis System, for time-value of money analysis of a variety of investments | 542 F Chapter 10 The COUNTREG Procedure 3L @0 E w fi wyi 0 X W i y - 0 - exp x00 1 a exp x00 1 1 _ exp z0 y 1 a exp xi 0 yi - exp xi 0 I _1 a exp xi 0 _ 1 xi 3L x a 2 1 a exp xi0 ln 1 a exp xi 0 - a exp xi0 3a i y og 1 exp zi y 1 a exp xi 0 1 V 1 a exp xi 0 C E w i yi 0 yi-1 1 a 2 X- 1 n j 0 j a 1 a 2 ln 1 a exp x0 0 yi - exp xi 0 a 1 a exp xi 0 ZINB Model with Standard Normal Link Function For this model the probability i is specified with the standard normal distribution function probit function i z y . The log-likelihood function is L X Wi ln j ziy 1 - ziy 1 a exp xi0 1J i yi 0 X Wi ln 1 - z y i Wyi 0 yi -1 X Wi X ln 7 a 1 i yi 0 j 0 - X Wi ln yi i yi 0 - X Wi yi a-1 ln 1 a exp x0 0 i yi 0 X Wi yi ln a i yi 0 X Wi yi x00 i yi 0 See Poisson Regression on page 534 for the definition of Wi. The gradient for this model is given by 9 2 ziy 1 - 1 aexp xi0 1 . Wi _ --------- ------ZT zi @y ziy 1 - ziy 1 aexp xi0 i yi 0 Computational Resources F 543 9L 9fl -Y. w wyi 0g z y 1 - .z i y z E y 0 w 1 - h z y exp x fl 1 C a exp xi fl 1 z y C 1 - z y 1 C a exp x fl - -1 1 C E w y 0 y - exp x . fl 1 C a exp xi fl 9L _ x 1 - zi y a 2 1 C a exp xifl ln 1 C a exp xi fl - a exp xi fl 9a 1 z y 1 C a exp xifl 1 V c 1 - h z y 1 C aexp xifl i Wyi 0 i i i i y -1 1 C E w i Wyi 0 -a 2 z 1 _n a 2 ln 1 a eXP xi fl eXP x n 7 0 j C a 1 a 1 C a exp xifl Computational Resources The time and memory required by PROC COUNTREG are proportional to the number of parameters in the model and the number of observations in the data set being analyzed. Less time and memory are required for smaller models and fewer observations. Also affecting these resources are the method chosen to calculate the variance-covariance matrix and the optimization method. All optimization methods available through the METHOD option have similar memory use requirements. The processing time might differ for each method depending on the number of iterations and functional calls needed. The data set is read into memory to save .
