Generalized linear regression model for prediction of interval between the interview moment and the last episode of risky behavior on the base of psychosocial and demographic traits

This paper studies association between time till last alcohol consumption (that can be used as an indicator of risky behavior rate) and socio-demographical and psychological characteristics of respondents. We construct the appropriate linear regression model with time till last alcohol consumption as response variable. We examine the model using such diagnostics as residuals plots, test for outliers test for normality and others.