This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.
|CE_11C||Mon, 8/2/2010, 8:00 AM - 12:00 PM||CC-8 (East)|
|Adaptive Analysis of Data: Tests of Significance and Confidence Intervals — Continuing Education Course|
|Instructor(s): Tom O'Gorman, Northern Illinois University|
|In this session we will present several adaptive methods for the analysis of data. We will begin with a two-sample adaptive test. We then will present an adaptive method of testing any subset of coefficients in a multiple regression model and will show how to perform the adaptive tests using R and SAS. Adaptive tests for interaction and main effects in the analysis of factorial experiments and adaptive tests for slope will be described. We will show, using simulation studies, that the adaptive tests usually are more powerful than the traditional tests for non-normal error distributions. Since there is very little power loss with normal error distributions, adaptive tests can be recommended for general use in studies having more than 20 observations. Adaptive tests that are used in the analysis of repeated measurements will be described and compared to the non-adaptive mixed model tests. In addition, we will describe a method of computing adaptive confidence intervals. The attendees should be familiar with basic statistical modeling, including multiple regression and the analysis of variance, at the level of Applied Regression Analysis (1998) by Draper and Smith.|