Dear translators,

The following strings, which are all found in the source file
lib/src/modelprint.c, are printed in the context of a fixed
tabular format and they should be restricted to a maximum of
22 characters.

This list can be checked or updated  by doing

  fgrep '/* 22: ' ../lib/src/modelprint.c | awk -F"N_" '{print $NF}'

("Mean dependent var"),  /* 22: Mean of dependent variable */
("S.D. dependent var"),  /* 22: Standard deviation of dependent var */
("Sum squared resid"),   /* 22: Sum of squared residuals */
("S.E. of regression"),  /* 22: Standard error of the regression */
("R-squared"),           /* 22: */
("Adjusted R-squared"),  /* 22: */
("P-value(F)"),          /* 22: P-value of F-statistic */
("Log-likelihood"),      /* 22: */
("Akaike criterion"),    /* 22: Akaike Information Criterion */
("Schwarz criterion"),   /* 22: Schwarz Bayesian Criterion */
("Hannan-Quinn"),        /* 22: Hannan-Quinn Criterion */
("rho"),                 /* 22: 1st-order autocorrelation coeff. */
("Durbin-Watson")        /* 22: Durbin-Watson statistic */
("Uncentered R-squared"); /* 22: */
("Centered R-squared");  /* 22: */
("McFadden R-squared");  /* 22: McFadden's pseudo-R-squared */
("Smallest eigenvalue"); /* 22: */
("Mean of innovations"); /* 22: Mean of ARMA innovations */
("S.D. of innovations"); /* 22: Std. dev. of ARMA innovations */
("Median depend. var");  /* 22: Median of dependent variable */
("Sum absolute resid");  /* 22: Sum of absolute residuals */
("Censored obs"); /* 22: Number of censored observations */

Note also that there is a special string in gretl.pot,

msgid "note on model statistics abbreviations here"

You may use this string to add a note regarding abbreviations
used in your translations of the short strings noted above.  If
you don't want to use this mechanism, leave the string
untranslated and it will not appear.