Mercurial > octave
comparison scripts/statistics/models/logistic_regression_derivatives.m @ 3426:f8dde1807dee
[project @ 2000-01-13 08:40:00 by jwe]
author | jwe |
---|---|
date | Thu, 13 Jan 2000 08:40:53 +0000 |
parents | 041ea33fbbf4 |
children | d8b731d3f7a3 |
comparison
equal
deleted
inserted
replaced
3425:8625164a0a39 | 3426:f8dde1807dee |
---|---|
1 ## Copyright (C) 1995, 1996, 1997 Kurt Hornik | 1 ## Copyright (C) 1995, 1996, 1997 Kurt Hornik |
2 ## | 2 ## |
3 ## This program is free software; you can redistribute it and/or modify | 3 ## This program is free software; you can redistribute it and/or modify |
4 ## it under the terms of the GNU General Public License as published by | 4 ## it under the terms of the GNU General Public License as published by |
5 ## the Free Software Foundation; either version 2, or (at your option) | 5 ## the Free Software Foundation; either version 2, or (at your option) |
6 ## any later version. | 6 ## any later version. |
7 ## | 7 ## |
8 ## This program is distributed in the hope that it will be useful, but | 8 ## This program is distributed in the hope that it will be useful, but |
9 ## WITHOUT ANY WARRANTY; without even the implied warranty of | 9 ## WITHOUT ANY WARRANTY; without even the implied warranty of |
10 ## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | 10 ## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
11 ## General Public License for more details. | 11 ## General Public License for more details. |
12 ## | 12 ## |
13 ## You should have received a copy of the GNU General Public License | 13 ## You should have received a copy of the GNU General Public License |
14 ## along with this file. If not, write to the Free Software Foundation, | 14 ## along with this file. If not, write to the Free Software Foundation, |
15 ## 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. | 15 ## 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. |
16 | 16 |
17 ## Called by logistic_regression. Calculates derivates of the | 17 ## Called by logistic_regression. Calculates derivates of the |
18 ## log-likelihood for ordinal logistic regression model. | 18 ## log-likelihood for ordinal logistic regression model. |
19 | 19 |
20 ## Author: Gordon K. Smyth <gks@maths.uq.oz.au> | 20 ## Author: Gordon K. Smyth <gks@maths.uq.oz.au> |
21 ## Adapted-By: KH <Kurt.Hornik@ci.tuwien.ac.at> | 21 ## Adapted-By: KH <Kurt.Hornik@ci.tuwien.ac.at> |
22 ## Description: Derivates of log-likelihood in logistic regression | 22 ## Description: Derivates of log-likelihood in logistic regression |
23 | 23 |
24 function [dl, d2l] ... | 24 function [dl, d2l] ... |
25 = logistic_regression_derivatives (x, z, z1, g, g1, p) | 25 = logistic_regression_derivatives (x, z, z1, g, g1, p) |
26 | 26 |
27 ## first derivative | 27 ## first derivative |
28 v = g .* (1 - g) ./ p; v1 = g1 .* (1 - g1) ./ p; | 28 v = g .* (1 - g) ./ p; v1 = g1 .* (1 - g1) ./ p; |
29 dlogp = [(dmult (v, z) - dmult (v1, z1)), (dmult (v - v1, x))]; | 29 dlogp = [(dmult (v, z) - dmult (v1, z1)), (dmult (v - v1, x))]; |
30 dl = sum (dlogp)'; | 30 dl = sum (dlogp)'; |
31 | 31 |
32 ## second derivative | 32 ## second derivative |
33 w = v .* (1 - 2 * g); w1 = v1 .* (1 - 2 * g1); | 33 w = v .* (1 - 2 * g); w1 = v1 .* (1 - 2 * g1); |
34 d2l = [z, x]' * dmult (w, [z, x]) - [z1, x]' * dmult (w1, [z1, x]) ... | 34 d2l = [z, x]' * dmult (w, [z, x]) - [z1, x]' * dmult (w1, [z1, x]) ... |
35 - dlogp' * dlogp; | 35 - dlogp' * dlogp; |
36 | 36 |
37 endfunction | 37 endfunction |