Grand Valley State University Associate Professor of Statistics
The goal of predictive modeling is to build a model or algorithm to estimate future occurrences of a target variable based on a set of input variables. There are two main types of error to be concerned about in predictive modeling, bias and variance. In this talk we will introduce basic ideas and terminology used in predictive modeling. We will then discuss the problem of over-fitting, the variance bias trade-off and some modern methods for model selection.