Regression Tools

I assume that you've all taken your basic statistics course(s) and therefore I won't go into the details of linear and non-linear regression by least-squares analysis. Instead, I'll focus on how to use MATLAB to perform the most common statistical analyses that you'll need in this course. To perform regression analyses in MATLAB there are three options:

1. The Curve Fitting App is an interactive feature of MATLAB that lets you play with various fitting procedures. It's quite helpful if you're not sure what form best represents your data or if you need to do some quick analyses and don't care about integration of analysis results into other functions. Generally the App can handle only one independent (predictor) variable.

2. The FIT function is the function underneath the hood of the Curve Fitting App and therefore has all the same functionality but you'll need to know how to use it and which functions it calls on to make various calculations. The examples on these pages show you many of the common functions you'll need for this course and therefore I recommend you use FIT instead of the App for most analyses. Generally the App can handle only one independent (predictor) variable.

3. The NLINFIT function performs nonlinear regression and is useful if you have more than one independent (predictor) variable. This will most likely come up only on the plate heat exchanger. If you have to use this I suggest you review the NLINFIT documentation as we won't be covering it here.