Extreme annual temperature of eighteen stations in Malaysia is fitted to the Generalized Extreme Value distribution. Stationary and non-stationary models with trend are considered for each station and the Likelihood Ratio test is used to determine the best-fitting model. Results show that three out of eighteen stations i.e. Bayan Lepas, Labuan and Subang favor a model which is linear in the location parameter. A hierarchical cluster analysis is employed to investigate the existence of similar behavior among the stations. Three distinct clusters are found in which one of them consists of the stations that favor the nonstationary model. T-year estimated return levels of the extreme temperature are provided based on the chosen models.
Binary outcomes which are observed repeatedly over time may be dependent not only on the covariates but also on each other. The repeated outcomes represent correlated binary longitudinal data and the modeling of such data was initiated by Liang and Zeger (1986) through the use of generalized estimating equations (GEE). Specification of the association structure among the outcomes remains a challenge associated with the GEE approach. This paper compares the GEE approach that uses correlation to the alternating logistic regression (ALR) approach that uses odds ratio, to model the association among outcomes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.