SAS macro (CP) was written and tested using a SAS®9.3 and is able to perform the maximization process on a subjective interval to find the change-point, hazard rates and some other useful parameters that are mentioned in a variety of related articles. For a functional performance, the following parameters are to be defined for the macro: i) dataset name (indate), ii) two variables for time and event status (0=censored, 1=failure), respectively, iii) the lower bound of the maximization interval(l), iv) the upper bound of the maximization interval(u), v) space distance between grid points (s). The maximum value of the log-likelihood function is found by a grid search method and selecting the number of grid points depends on your computer’s RAM.
Working Paper (2021)
Fernández-de-Marcos Alberto and García-Portugués Eduardo