Spline Regression Techniques for Trend Detection

Document Type

Presentation Abstract

Presentation Date

3-8-2007

Abstract

Trend estimation is important in many fields, though arguably, the most important applications appear in ecology. Trend is difficult to quantify, in fact, the term itself is not well-defined. Commonly, trend is quantified by estimating the slope coefficient in a regression model where the response variable is an index of population size, and time is the explanatory variable. Linear trend is often unrealistic for biological populations; in fact, many critical environmental changes occur abruptly as a result of very rapid changes in human activities. My PhD research has involved formulating methods with greater flexibility and robustness than those currently in use. This is a particularly timely and important problem given rapid changes in the environment associated with change in land use, human pressures, and global warming.

Penalized spline regression provides a flexible technique for fitting a smooth curve. This method has proven useful in many areas including environmental monitoring, however, inference is more difficult than ordinary linear regression because so many parameters are estimated. My research focuses on developing a test for trend using penalized spline regression techniques. The test will be compared to other commonly used methods in trend detection such as linear regression, nonparametric correlation techniques, and other smoothing techniques. Further considerations such as autocorrelation are considered. In this presentation, penalized spline regression will be introduced and possible trend detection methods using spline regression will be discussed. Simulation results and future research will also be presented.

Additional Details

Thursday, 8 March 2007
4:10 p.m. in Math 109

This document is currently not available here.

Share

COinS