Home > chronux > locfit > m > residuals.m

# residuals

## PURPOSE

Residuals (or a few other things) from a locfit() fit.

## SYNOPSIS

function y = residuals(fit,type)

## DESCRIPTION

``` Residuals (or a few other things) from a locfit() fit.

Input arguments:
fit - the locfit() fit.
type (optional) type of residuals. Valid types are
'dev'    (deviance, the default)
'd2'     (deviance squared)
'pearson'(Pearson)
'raw'    (observed - fitted)
'ldot'   (derivative of log-likelihood)
'lddot'  (second derivative)
'fit'    (fitted values - no transformation)
'mean'   (fitted values - with back transformation)

## CROSS-REFERENCE INFORMATION

This function calls:
• predict Interpolate a fit produced by locfit().
This function is called by:

## SOURCE CODE

```0001 function y = residuals(fit,type)
0002
0003 % Residuals (or a few other things) from a locfit() fit.
0004 %
0005 % Input arguments:
0006 %   fit - the locfit() fit.
0007 %   type (optional) type of residuals. Valid types are
0008 %     'dev'    (deviance, the default)
0009 %     'd2'     (deviance squared)
0010 %     'pearson'(Pearson)
0011 %     'raw'    (observed - fitted)
0012 %     'ldot'   (derivative of log-likelihood)
0013 %     'lddot'  (second derivative)
0014 %     'fit'    (fitted values - no transformation)
0015 %     'mean'   (fitted values - with back transformation)
0016 %
0017 %  Author: Catherine Loader.
0018
0019 if (nargin<2) type = 'dev'; end;
0020
0021 y = predict(fit,'d','restyp',type);
0022
0023 return;```

Generated on Fri 28-Sep-2012 12:34:30 by m2html © 2005