User`s guide

Additional Preprocessing Steps

2-41

There are several disadvantages associated with performing transformations:

• For the log transformation, negative response values cannot be processed.

• For all transformations, the basic assumption that the residual variance is

constant is violated. To avoid this problem, you could plot the residuals on

the transformed scale. For the power transformation shown above, the

transformed scale is given by the residuals

Note that the residual plot associated with the Curve Fitting Tool does not

support transformed scales.

Deciding on a particular transformation is not always obvious. However, a

scatter plot will often reveal the best form to use. In practice you can

experiment with various transforms and then plot the residuals from the

command line using the transformed scale. If the errors are reasonable (they

appear random with minimal scatter, and don’t exhibit any systematic

behavior), the transform is a good candidate.

Removing Infs, NaNs, and Outliers

Although the Curve Fitting Toolbox ignores Infs and NaNs when fitting data,

and you can exclude outliers during the fitting process, you might still want to

remove this data from your data set. To do so, you modify the associated data

set variables from the MATLAB command line.

For example, when using toolbox functions such as

fit from the command line,

you must supply predictor and response vectors that contain finite numbers. To

remove

Infs, you can use the isinf function.

ind = find(isinf(xx));

xx(ind) = [];

yy(ind) = [];

To remove NaNs, you can use the isnan function. For examples that remove

NaNs and outliers from a data set, refer to “Data Preprocessing” in the MATLAB

documentation.

1–

–=