1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
|
/*
* gretl -- Gnu Regression, Econometrics and Time-series Library
* Copyright (C) 2001 Allin Cottrell and Riccardo "Jack" Lucchetti
*
* 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 "libgretl.h"
#include "version.h"
#include "libset.h"
static int get_range_and_mean (int t1, int t2, const double *x,
double *range, double *mean,
gretlopt opt)
{
double xsum = 0.0, xmin = x[t1], xmax = x[t1];
int t, n = 0;
int err = 0;
if (opt & OPT_T) {
/* trim the extrema */
int T = t2 - t1 + 1;
double *xs = malloc(T * sizeof *xs);
if (xs == NULL) {
err = E_ALLOC;
} else {
int s = 0;
for (t=t1; t<=t2; t++) {
if (!na(x[t])) {
xs[s++] = x[t];
n++;
}
}
if (n < 4) {
err = E_TOOFEW;
n = 0;
} else {
qsort(xs, n, sizeof *x, gretl_compare_doubles);
xmin = xs[1];
xmax = xs[n-2];
for (s=1; s<n-1; s++) {
xsum += xs[s];
}
n -= 2;
}
free(xs);
}
} else {
/* no trimming: use all available data */
for (t=t1; t<=t2; t++) {
if (!na(x[t])) {
xmax = (x[t] > xmax) ? x[t] : xmax;
xmin = (x[t] < xmin) ? x[t] : xmin;
xsum += x[t];
n++;
}
}
}
if (n > 0) {
*mean = xsum / n;
*range = xmax - xmin;
} else {
*mean = NADBL;
*range = NADBL;
if (!err) {
err = E_DATA;
}
}
return err;
}
static int do_range_mean_plot (const gretl_matrix *y,
const gretl_matrix *X,
double b0, double b1,
const char *vname)
{
FILE *fp;
int fitline = 0;
int t, err = 0;
fp = open_plot_input_file(PLOT_RANGE_MEAN, 0, &err);
if (err) {
return err;
}
if (!na(b0) && !na(b1)) {
fitline = 1;
}
fprintf(fp, "# for %s\n", vname);
fputs("set nokey\n", fp);
fprintf(fp, "set title '%s %s %s'\n",
_("range-mean plot for"), vname,
(fitline)? _("with least squares fit") : "");
fprintf(fp, "set xlabel '%s'\nset ylabel '%s'\n",
_("mean"), _("range"));
fputs("plot \\\n", fp);
gretl_push_c_numeric_locale();
if (fitline) {
fprintf(fp, "%g+%g*x notitle w lines lt 2 ,\\\n", b0, b1);
}
fputs("'-' using 1:2 w points lt 1\n", fp);
for (t=0; t<X->rows; t++) {
fprintf(fp, "%g %g\n", gretl_matrix_get(X, t, 1),
gretl_vector_get(y, t));
}
fputs("e\n", fp);
gretl_pop_c_numeric_locale();
return finalize_plot_input_file(fp);
}
/* drop first/last observations from sample if missing obs
encountered */
static int
rm_adjust_sample (int v, const DATASET *dset, int *t1, int *t2)
{
int t, t1min = *t1, t2max = *t2;
for (t=t1min; t<t2max; t++) {
if (na(dset->Z[v][t])) t1min++;
else break;
}
for (t=t2max; t>t1min; t--) {
if (na(dset->Z[v][t])) t2max--;
else break;
}
*t1 = t1min;
*t2 = t2max;
return 0;
}
static int get_n_samples (int T, int pd, int mmin)
{
int k = (int) sqrt((double) T);
if (k < mmin) {
k = mmin;
} else {
/* If pd >= mmin and is "not too far" from k, it may make sense to
use pd instead of sqrt(T) as the subsample length. This will,
e.g., make each subsample a year, with quarterly or monthly
data.
*/
if (pd >= mmin && T/pd >= 4 &&
pd >= 2.0*k/3.0 && pd <= 3.0*k/2.0) {
k = pd;
}
}
return k;
}
int range_mean_graph (int vnum, DATASET *dset,
gretlopt opt, PRN *prn)
{
gretl_matrix_block *B;
gretl_matrix *y, *X, *b, *V;
int k, t, m, T, mmin, rem;
int quiet = (opt & OPT_Q);
char startdate[OBSLEN], enddate[OBSLEN];
int t1 = dset->t1;
int t2 = dset->t2;
int digits;
int err = 0;
rm_adjust_sample(vnum, dset, &t1, &t2);
T = t2 - t1 + 1;
/* We need at least 4 sub-samples to do this, and each
must have at least @mmin observations, the specific
value depending on whether or not we're trimming the
extrema in each sub-sample.
*/
mmin = (opt & OPT_T)? 6 : 4;
if (T < 4 * mmin) {
pputs(prn, _("Sample is too small for range-mean graph\n"));
return E_TOOFEW;
}
k = get_n_samples(T, dset->pd, mmin);
rem = T % k;
m = (T / k) + ((rem >= mmin)? 1 : 0);
B = gretl_matrix_block_new(&y, m, 1,
&X, m, 2,
&b, 2, 1,
&V, 2, 2,
NULL);
if (B == NULL) {
return E_ALLOC;
}
if (!quiet) {
pprintf(prn, _("Range-mean statistics for %s\n"),
dset->varname[vnum]);
pprintf(prn, _("using %d sub-samples of size %d\n\n"),
m, k);
pprintf(prn, "%30s%16s\n", _("range"), _("mean"));
}
digits = get_gretl_digits();
/* find sub-sample means and ranges */
for (t=0; t<m; t++) {
int start = t1 + t * k;
int end = start + k - 1;
double range, mean;
if (end > t2) {
end = t2;
} else if (t2 - end <= rem && rem < mmin) {
end += rem;
}
err = get_range_and_mean(start, end, dset->Z[vnum],
&range, &mean, opt);
if (err) {
break;
}
gretl_vector_set(y, t, range);
gretl_matrix_set(X, t, 0, 1.0);
gretl_matrix_set(X, t, 1, mean);
if (!quiet) {
int len;
ntolabel(startdate, start, dset);
ntolabel(enddate, end, dset);
len = pprintf(prn, "%s - %s", startdate, enddate);
bufspace(20 - len, prn);
gretl_print_fullwidth_double(range, digits, prn);
gretl_print_fullwidth_double(mean, digits, prn);
pputc(prn, '\n');
}
}
if (!err) {
double b0 = NADBL, b1 = NADBL;
double s2;
err = gretl_matrix_ols(y, X, b, V, NULL, &s2);
if (err) {
pputs(prn, _("Error estimating range-mean model\n"));
errmsg(err, prn);
} else {
double bse = sqrt(gretl_matrix_get(V, 1, 1));
double tstat, pv;
if (!quiet) {
pputc(prn, '\n');
pprintf(prn, _("slope of range against mean = %g\n"),
b->val[1]);
}
if (bse > 0) {
tstat = b->val[1] / bse;
pv = student_pvalue_2(m - 2, tstat);
record_test_result(tstat, pv);
if (!quiet) {
pprintf(prn, _("p-value for H0: slope = 0 is %g\n"), pv);
}
if (pv < .10) {
/* arrange for fitted line to be drawn */
b0 = b->val[0];
b1 = b->val[1];
}
}
}
if (!err && !quiet) {
if (gnuplot_graph_wanted(PLOT_RANGE_MEAN, opt, &err)) {
err = do_range_mean_plot(y, X, b0, b1, dset->varname[vnum]);
}
}
}
gretl_matrix_block_destroy(B);
return err;
}
|