The ability to predict the arrival and residence time of mobile users at a particular place is essential for the development of a wealth of new applications and services, such as smart heating control, transportation planning or urban navigation. Previous techniques based on probabilistic models have not been able to perform such prediction accurately. In this paper, we present two linear mobility models, namely Linear Regression, and Auto-Regression, to predict the temporal behavior, particularly the residence time, of individual users. We run performance evaluation experiments on two different WiFi mobility traces datasets made available through the CRAWDAD project. Our results show that using linear regression-based learning algorithms significantly improve the residence time prediction accuracy compared to state-of-the-art methods, and achieve prediction errors in the order of seconds and minutes for a large number of users.