Thesis (Ph.D.) - Cairo University - Institute of Statistical Studies and Research - Department of Statistics and Econometrics

A new regression model for doubly bounded continuous dependent variables (DBCDVs) such as proportions, ratios, and rates is introduced. The new model consists of two interdependent submodels for mean and precision and it helps to take account for the information in the DBCDVs properly. It is based on the unit gamma distribution. Expressions for the score vector and for fisher's information matrix are derived. The suggested model is studied for practical purposes using two applications. The results reveal that the new model may serve as an alternative to the beta and the simplex regression models that have appeared in the literature. These results were supported through a simulation study

Issued also as CD