XploRe Help : logsplinesummary

Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Library:	smoother
See also:	plogspline qlogspline dlogspline rlogspline logsplinefit plotlogspline

Quantlet:		logsplinesummary
Description:		summarizes the results of estimation of quantlet logsplinefit

Usage:	`summary = logsplinesummary(fit)`
Input:
	`fit`	a list like the output from logsplinefit
Output:
	`summary`	string vector containing the summary

Note:

This function returns a string output. The main body is a table with five columns: the first column contains a possible number of knots for the fitted model; the second column the log-likelihood for the fit; the third column represents -2*loglikelihood + penalty*(number of knots-1), which is the AIC criterion - logsplinefit selected the model with the smallest value of AIC; the fourth and fifth columns give the endpoints of the interval of values of penalty that would yield the model with the indicated number of knots. (NaN's imply that the model is not optimal for any choice of penalty.) At the bottom of the table the number of knots corresponding to the selected model is reported, as is the value of penalty that was used.

Example:

library("smoother") data = read("geyser") fit = logsplinefit(data[,1], "no","no","no","no","no",1,"no","no",7) logsplinesummary(fit)

Result:

A summary of the output of the logsplinefit quantlet:

Contents of summary

[ 1,] "=============================================================="
[ 2,] "                     LogSpline summary                        "
[ 3,] "--------------------------------------------------------------"
[ 4,] " knots  loglik      AIC           penalty                     "
[ 5,] "                                min     max                   "
[ 6,] "   4   -422.73   873.469    309.29      +INF                  "
[ 7,] "   5   -268.09   571.180     15.74    309.29                  "
[ 8,] "   6   -260.22   562.437      2.09     15.74                  "
[ 9,] "   7   -260.14   569.282      +NAN      +NAN                  "
[10,] "   8   -258.12   572.249      0.36      2.09                  "
[11,] "   9   -257.95   578.891      0.31      0.36                  "
[12,] "  10   -257.79   585.577      0.15      0.31                  "
[13,] "  11   -257.71   592.430      0.00      0.15                  "
[14,] "  12   -257.71   599.425      0.00      0.00                  "
[15,] "  13   -257.71   606.425      0.00      0.00                  "
[16,] "--------------------------------------------------------------"
[17,] " the present optimal number of knots is 6                     "
[18,] "--------------------------------------------------------------"
[19,] " penalty(AIC) was 7                                           "
[20,] " the default (BIC) would have been 5.606                      "
[21,] "--------------------------------------------------------------"
[22,] " models with fewer than 4 knots can be fitted, but they are   "
[23,] " not optimal for the present choice of penalty - choose       "
[24,] " penalty in logsplinefit larger to see these fits             "
[25,] "=============================================================="

Author: K. Komorad, W. Haerdle, 20010625 license MD*Tech