Category: Nonlinear regression. These pages describe regression models where you specify a nonlinear functional relationship. Articles are arranged by date with the most recent entries at the top. You can find the theme and closely related categories and other resources at the bottom of this page.

Stats: Fitting a difference in exponential functions (January 26, 2007). I was presented with a data set of sixteen subjects that showed a rise in values from zero to a maximum followed by gradual return of those values to zero for each subject. Data of this form can often be modeled by a difference of exponential functions.

Stats: (Seminar notes) Missing values in a dose response model (July 17, 2006). One of the talks at the 18th Annual Applied Statistics in Agriculture Conference, sponsored by Kansas State University was "Dose-Response Modeling with Marginal Information on Missing Categorical Covariate" by John R. Stevens, Utah State University. David I. Schlipalius, of The University of Queensland was a co-author.

Stats: Growth curves (March 1, 2005). The New York Times has a nice article about Dr. James Tanner, an expert on childhood growth. This article contrasts the growth charts developed by Dr. Tanner with growth charts developed by the U.S. Centers for Disease Control and Prevention.

Stats: Nonlinear Least Squares in S-plus and R (June 28, 2004). I've worked on a brief explanation of how to fit an S-shaped curve using SPSS and someone wanted to know how to do this in S-plus. Here's a simple example.

Stats: Guidelines for nonlinear regression models (May 26, 2004). There are three steps in a typical nonlinear regression analysis. [incomplete]

Stats: S-shaped curves (February 12, 2004). Competitive binding experiments will often need a nonlinear regression model. This model has to level off at both extremes to represent almost no binding at one end and saturated binding at the other end. This behavior is usually represented by an S-shaped curve. In this web page, I will describe some of the equations that you might use to represent an S-shaped curve.

Stats: A simple model of nonlinear growth (October 1, 2003). Part of the challenge in nonlinear regression is to choose the correct form of the nonlinear relationship. Part of this is intuition, part of this is understanding some basic mathematics, and part of it is just trial and error.

Theme and closely related categories:

Other resources:

[Return to full topic list] [Read current weblog entries]

This webpage was written by Steve Simon on 2007-08-10, edited by Steve Simon, and was last modified on 2008-07-08. Send feedback to ssimon at cmh dot edu or click on the email link at the top of the page.