In many longitudinal studies, nominal and ordinal mixed bivariate responses are measured. In these studies, the aim is to investigate the effects of explanatory variables on these time-related responses. A regression analysis for these types of data must allow for the correlation among responses during the time. To analyze such ordinal-nominal responses, using a proposed weighting approach, an ordinal and nominal mixed transition model is proposed and then maximum likelihood method is used to find the parameter estimates. The likelihood function in this method is partitioned to make possible the use of existing software.
Social-economical and political consequences arising from Iranian unemployment in the community and necessity of familiarity with the labor force characteristics particularly identification of the structure of changes in economic activity status of the Iranian population are important in order to achieve the objectives of social-economic and cultural development plans of the country. Data of Labor Force Survey in Iran are in a longitudinal form... [To continue please click here]
Rezaei Ghahroodi Z, Ganjali M, Harandi F. Transition Models for Analyzing Longitudinal Data with Bivariate Mixed Ordinal and Nominal Responses . JSRI 2008; 5 (1) :75-94 URL: http://jsri.srtc.ac.ir/article-1-127-en.html