Added variable/parameter estimate matching via new struct

diff --git a/lib/linreg/linreg.c b/lib/linreg/linreg.c
index 26efe886c7497725862dec090bf087227b75ca8e..85b302d188e2f01b215a805ddd3d6659c4dd2a7d 100644 (file)
--- a/lib/linreg/linreg.c
+++ b/lib/linreg/linreg.c
@@ -87,38 +87,39 @@ linreg_mean_std (gsl_vector_const_view v, double *mp, double *sp, double *ssp)
 pspp_linreg_cache *
 pspp_linreg_cache_alloc (size_t n, size_t p)
 {
-  pspp_linreg_cache *cache;
+  pspp_linreg_cache *c;
 
-  cache = (pspp_linreg_cache *) malloc (sizeof (pspp_linreg_cache));
-  cache->param_estimates = gsl_vector_alloc (p + 1);
-  cache->indep_means = gsl_vector_alloc (p);
-  cache->indep_std = gsl_vector_alloc (p);
-  cache->ssx = gsl_vector_alloc (p);   /* Sums of squares for the independent
+  c = (pspp_linreg_cache *) malloc (sizeof (pspp_linreg_cache));
+  c->param_estimates = gsl_vector_alloc (p + 1);
+  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.
                                         */
-  cache->ss_indeps = gsl_vector_alloc (p);     /* Sums of squares for the model 
-                                                  parameters. 
-                                                */
-  cache->cov = gsl_matrix_alloc (p + 1, p + 1);        /* Covariance matrix. */
-  cache->n_obs = n;
-  cache->n_indeps = p;
+  c->ss_indeps = gsl_vector_alloc (p); /* Sums of squares for the model 
+                                          parameters. 
+                                        */
+  c->cov = gsl_matrix_alloc (p + 1, p + 1);    /* Covariance matrix. */
+  c->n_obs = n;
+  c->n_indeps = p;
   /*
      Default settings.
    */
-  cache->method = PSPP_LINREG_SWEEP;
+  c->method = PSPP_LINREG_SWEEP;
 
-  return cache;
+  return c;
 }
 
 void
-pspp_linreg_cache_free (pspp_linreg_cache * cache)
+pspp_linreg_cache_free (pspp_linreg_cache * c)
 {
-  gsl_vector_free (cache->param_estimates);
-  gsl_vector_free (cache->indep_means);
-  gsl_vector_free (cache->indep_std);
-  gsl_vector_free (cache->ss_indeps);
-  gsl_matrix_free (cache->cov);
-  free (cache);
+  gsl_vector_free (c->param_estimates);
+  gsl_vector_free (c->indep_means);
+  gsl_vector_free (c->indep_std);
+  gsl_vector_free (c->ss_indeps);
+  gsl_matrix_free (c->cov);
+  free (c->coeff);
+  free (c);
 }
 
 /*
@@ -180,9 +181,9 @@ pspp_linreg (const gsl_vector * Y, const gsl_matrix * X,
          standard deviations of the independent variables here since doing
          so would cause a miscalculation of the residual sums of
          squares. Dividing by the standard deviation is done GSL's linear
-         regression functions, so if the design matrix has a very poor
+         regression functions, so if the design matrix has a poor
          condition, use QR decomposition.
-         *
+
          The design matrix here does not include a column for the intercept
          (i.e., a column of 1's). If using PSPP_LINREG_QR, we need that column,
          so design is allocated here when sweeping, or below if using QR.
@@ -244,6 +245,7 @@ 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;
          gsl_vector_set (cache->param_estimates, i + 1, tmp);
          m -= tmp * gsl_vector_get (cache->indep_means, i);
        }

diff --git a/lib/linreg/pspp_linreg.h b/lib/linreg/pspp_linreg.h
index 5fa4e63bc45c06c4166e741bb5e6f9c01825bea0..40c5ad842b3df324c071d3d4771ffd6dc6cd39e4 100644 (file)
--- a/lib/linreg/pspp_linreg.h
+++ b/lib/linreg/pspp_linreg.h
@@ -63,23 +63,36 @@ enum
 
 /*
   Cache for the relevant data from the model. There are several
-  members which the caller may not use, and which could use a lot of
+  members which the caller might not use, and which could use a lot of
   storage. Therefore non-essential members of the struct will be
   allocated only when requested.
  */
+struct pspp_linreg_coeff
+{
+  double estimate; /* Estimated coefficient. */
+  const struct variable *v; /* The variable associated with this coefficient. 
+                              The calling function should supply the variable
+                              when it creates the design matrix. The estimation
+                              procedure ignores the struct variable *. It is here so
+                              the caller can match parameters with relevant 
+                              variables.
+                           */
+};
 struct pspp_linreg_cache_struct
 {
   int n_obs;                   /* Number of observations. */
   int n_indeps;                        /* Number of independent variables. */
+  int n_coeffs;
 
   /* 
-     The var structs are ignored during estimation.
-     They are here so the calling procedures can
-     find the variables used in the model.
+     The variable struct is ignored during estimation.
+     It is here so the calling procedure can
+     find the variable used in the model.
   */
-  struct var *depvar;
-  struct var **indepvar;   
+  const struct variable *depvar;
+
   gsl_vector *residuals;
+  struct pspp_linreg_coeff *coeff;
   gsl_vector *param_estimates;
   int method;                  /* Method to use to estimate parameters. */
   /*