Home | Main | Session 1 | Session 2 | Session 3

Are Mixed Models Useful?

According to Google (about 43,700,000 results), a mixed model is “a statistical model containing both fixed effects and random effects. These models are useful in a wide variety of disciplines in the physical, biological and social sciences”.

And mixed models are currently very popular. In fact Google gives only about 14,300,000 results for linear models, which are said to “describe a continuous response variable as a function of one or more predictor variables”.

However, a problem with mixed models is that by including random effects their results are not easily extended to other populations.

This workshop will be held in the computer laboratory of Statistics Department at Mae Jo University in Chiang Mai from 8.30 am to 4.30 pm on Friday July 22 2016 on the 5th floor of the Science Faculty building. We will give examples of mixed models in medical and biological science. Outcome (response) variables include growth stunting among infants in Nepal (a continuous outcome), zooplankton and fingerling densities in an estuary in Southern Thailand (both continuous and binary outcomes), and leaf monkey behaviour in a small peninsula in the Gulf of Thailand (binary outcomes). In each case we will compare classical linear and logistic models with corresponding mixed models. Using informative graphical displays, results will be compared with the aim of understanding advantages and disadvantages of mixed models.

Dept of Mathematics & Computer Science, Faculty of Science & Technology, PSU Faculties of Science, Mae Jo University

Last updated: July 22, 2016