Mercurial > hg > ltpda

comparison m-toolbox/classes/+utils/@math/wfun.m @ 0:f0afece42f48

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Import.
author Daniele Nicolodi <nicolodi@science.unitn.it>
date Wed, 23 Nov 2011 19:22:13 +0100
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:f0afece42f48
1 % WFUN defines weighting factor for fitting procedures ctfit, dtfit.
2 %
3 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4 %
5 % DESCRIPTION
6 %
7 % Defines the weigthing factor for the fitting procedure performed by
8 % ctfit and dtfit.
9 %
10 % CALL:
11 %
12 % weight = wfun(y,weightparam)
13 %
14 % INPUT:
15 %
16 % y: are the set of data to be fitted
17 % weightparam: is a parameter for the swhitching procedure. Admitted
18 % values are:
19 % weightparam = 1 --> equal weights (one) for each point. This is the
20 % default option.
21 % weightparam = 2 --> weight with the inverse of absolute value of
22 % data
23 % weightparam = 3 --> weight with square root of the inverse of
24 % absolute value of data
25 % weightparam = 4 --> weight with the inverse of the square mean
26 % spread
27 %
28 %
29 % OUTPUT:
30 %
31 % weight: is the set of weighting factors
32 %
33 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34 % VERSION: $Id: wfun.m,v 1.5 2008/11/17 16:52:23 luigi Exp $
35 %
36 % HISTORY: 08-10-2008 L Ferraioli
37 % Creation
38 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
39
40 function weight = wfun(y,weightparam)
41
42 [a,b] = size(y);
43
44 opt = 0; % default value
45 if ~isempty(weightparam)
46 opt = weightparam;
47 end
48
49 switch opt
50 case 0
51 disp(' Using external weights... ')
52 case 1
53 weight = ones(a,b); % equal weights for each point
54 case 2
55 weight = 1./abs(y); % weight with the inverse of absolute value of data
56 case 3
57 weight = 1./sqrt(abs(y)); % weight with square of the inverse of absolute value of data
58 case 4
59 my = mean(y,max(a,b));
60 weight = 1./((y-my).^2); % weight with the inverse of the square mean spread
61 end