X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fmath%2Flinreg%2Flinreg.c;h=6bd450a1a9b3ecfe3833e0e11e6015ea43e83665;hb=2da08e801be698df268f35ac6e21cc8a345ba1fb;hp=2945b26d87002f9983d1267d041ca5433ce6ac25;hpb=6e0bcaf730de6ad07b874cd02fba32ccfb2a6f47;p=pspp-builds.git

diff --git a/src/math/linreg/linreg.c b/src/math/linreg/linreg.c
index 2945b26d..6bd450a1 100644
--- a/src/math/linreg/linreg.c
+++ b/src/math/linreg/linreg.c
@@ -1,23 +1,20 @@
-/*
-  lib/linreg/linreg.c
-  
-  Copyright (C) 2005 Free Software Foundation, Inc. Written by Jason H. Stover.
-  
-  This program is free software; you can redistribute it and/or modify it under
-  the terms of the GNU General Public License as published by the Free
-  Software Foundation; either version 2 of the License, or (at your option)
-  any later version.
-  
-  This program is distributed in the hope that it will be useful, but WITHOUT
-  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
-  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
-  more details.
-  
-  You should have received a copy of the GNU General Public License along with
-  this program; if not, write to the Free Software Foundation, Inc., 51
-  Franklin Street, Fifth Floor, Boston, MA 02111-1307, USA.
- */
+/* PSPP - a program for statistical analysis.
+   Copyright (C) 2005 Free Software Foundation, Inc. 
+
+   This program is free software: you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation, either version 3 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program.  If not, see <http://www.gnu.org/licenses/>. */
 
+#include <config.h>
 #include <gsl/gsl_fit.h>
 #include <gsl/gsl_multifit.h>
 
@@ -53,7 +50,7 @@
 */
 
 #include <math/linreg/linreg.h>
-#include <math/linreg/coefficient.h>
+#include <math/coefficient.h>
 #include <gsl/gsl_errno.h>
 #include <linreg/sweep.h>
 /*
@@ -88,6 +85,54 @@ linreg_mean_std (gsl_vector_const_view v, double *mp, double *sp, double *ssp)
   return GSL_SUCCESS;
 }
 
+/*
+  Set V to contain an array of pointers to the variables
+  used in the model. V must be at least C->N_COEFFS in length.
+  The return value is the number of distinct variables found.
+ */
+int
+pspp_linreg_get_vars (const void *c_, const struct variable **v)
+{
+  const pspp_linreg_cache *c = c_;
+  struct pspp_coeff *coef = NULL;
+  const struct variable *tmp;
+  int i;
+  int result = 0;
+
+  /*
+     Make sure the caller doesn't try to sneak a variable
+     into V that is not in the model.
+   */
+  for (i = 0; i < c->n_coeffs; i++)
+    {
+      v[i] = NULL;
+    }
+  /*
+     Start at c->coeff[1] to avoid the intercept.
+   */
+  v[result] = pspp_coeff_get_var (c->coeff[1], 0);
+  result = (v[result] == NULL) ? 0 : 1;
+
+  for (coef = c->coeff[2]; coef < c->coeff[c->n_coeffs]; coef++)
+    {
+      tmp = pspp_coeff_get_var (coef, 0);
+      assert (tmp != NULL);
+      /* Repeated variables are likely to bunch together, at the end
+         of the array. */
+      i = result - 1;
+      while (i >= 0 && v[i] != tmp)
+	{
+	  i--;
+	}
+      if (i < 0 && result < c->n_coeffs)
+	{
+	  v[result] = tmp;
+	  result++;
+	}
+    }
+  return result;
+}
+
 /*
   Allocate a pspp_linreg_cache and return a pointer
   to it. n is the number of cases, p is the number of
@@ -103,11 +148,11 @@ pspp_linreg_cache_alloc (size_t n, size_t p)
   c->indep_means = gsl_vector_alloc (p);
   c->indep_std = gsl_vector_alloc (p);
   c->ssx = gsl_vector_alloc (p);	/* Sums of squares for the
-					   independent variables. 
-					*/
+					   independent variables.
+					 */
   c->ss_indeps = gsl_vector_alloc (p);	/* Sums of squares for the
-					   model parameters. 
-					*/
+					   model parameters.
+					 */
   c->cov = gsl_matrix_alloc (p + 1, p + 1);	/* Covariance matrix. */
   c->n_obs = n;
   c->n_indeps = p;
@@ -116,22 +161,37 @@ pspp_linreg_cache_alloc (size_t n, size_t p)
    */
   c->method = PSPP_LINREG_SWEEP;
   c->predict = pspp_linreg_predict;
-  c->residual = pspp_linreg_residual; /* The procedure to comput my residuals. */
-  c->resid = NULL; /* The variable storing my residuals. */
+  c->residual = pspp_linreg_residual;	/* The procedure to compute my
+					   residuals. */
+  c->get_vars = pspp_linreg_get_vars;	/* The procedure that returns
+					   pointers to model
+					   variables. */
+  c->resid = NULL;		/* The variable storing my residuals. */
+  c->pred = NULL;		/* The variable storing my predicted values. */
 
   return c;
 }
 
 bool
-pspp_linreg_cache_free (void * m)
+pspp_linreg_cache_free (void *m)
 {
+  int i;
+
   pspp_linreg_cache *c = m;
-  gsl_vector_free (c->indep_means);
-  gsl_vector_free (c->indep_std);
-  gsl_vector_free (c->ss_indeps);
-  gsl_matrix_free (c->cov);
-  pspp_linreg_coeff_free (c->coeff);
-  free (c);
+  if (c != NULL)
+    {
+      gsl_vector_free (c->indep_means);
+      gsl_vector_free (c->indep_std);
+      gsl_vector_free (c->ss_indeps);
+      gsl_matrix_free (c->cov);
+      gsl_vector_free (c->ssx);
+      for (i = 0; i < c->n_coeffs; i++)
+	{
+	  pspp_coeff_free (c->coeff[i]);
+	}
+      free (c->coeff);
+      free (c);
+    }
   return true;
 }
 
@@ -145,7 +205,7 @@ pspp_linreg (const gsl_vector * Y, const gsl_matrix * X,
 	     const pspp_linreg_opts * opts, pspp_linreg_cache * cache)
 {
   int rc;
-  gsl_matrix *design;
+  gsl_matrix *design = NULL;
   gsl_matrix_view xtx;
   gsl_matrix_view xm;
   gsl_matrix_view xmxtx;
@@ -187,8 +247,8 @@ pspp_linreg (const gsl_vector * Y, const gsl_matrix * X,
   cache->dfm = cache->n_indeps;
   cache->dfe = cache->dft - cache->dfm;
   cache->n_coeffs = X->size2 + 1;	/* Adjust this later to allow for
-					   regression through the origin. 
-					*/
+					   regression through the origin.
+					 */
   if (cache->method == PSPP_LINREG_SWEEP)
     {
       gsl_matrix *sw;
@@ -262,7 +322,7 @@ pspp_linreg (const gsl_vector * Y, const gsl_matrix * X,
       for (i = 0; i < cache->n_indeps; i++)
 	{
 	  tmp = gsl_matrix_get (sw, i, cache->n_indeps);
-	  cache->coeff[i + 1].estimate = tmp;
+	  cache->coeff[i + 1]->estimate = tmp;
 	  m -= tmp * gsl_vector_get (cache->indep_means, i);
 	}
       /*
@@ -298,7 +358,7 @@ pspp_linreg (const gsl_vector * Y, const gsl_matrix * X,
 	    }
 	  gsl_matrix_set (cache->cov, 0, 0, tmp);
 
-	  cache->coeff[0].estimate = m;
+	  cache->coeff[0]->estimate = m;
 	}
       else
 	{
@@ -308,6 +368,18 @@ pspp_linreg (const gsl_vector * Y, const gsl_matrix * X,
 	}
       gsl_matrix_free (sw);
     }
+  else if (cache->method == PSPP_LINREG_CONDITIONAL_INVERSE)
+    {
+      /*
+	Use the SVD of X^T X to find a conditional inverse of X^TX. If
+	the SVD is X^T X = U D V^T, then set the conditional inverse
+	to (X^T X)^c = V D^- U^T. D^- is defined as follows: If entry
+	(i, i) has value sigma_i, then entry (i, i) of D^- is 1 /
+	sigma_i if sigma_i > 0, and 0 otherwise. Then solve the normal
+	equations by setting the estimated parameter vector to 
+	(X^TX)^c X^T Y.
+       */
+    }
   else
     {
       gsl_multifit_linear_workspace *wk;
@@ -333,7 +405,7 @@ pspp_linreg (const gsl_vector * Y, const gsl_matrix * X,
 				cache->cov, &(cache->sse), wk);
       for (i = 0; i < cache->n_coeffs; i++)
 	{
-	  cache->coeff[i].estimate = gsl_vector_get (param_estimates, i);
+	  cache->coeff[i]->estimate = gsl_vector_get (param_estimates, i);
 	}
       if (rc == GSL_SUCCESS)
 	{