Nonparametric Methods in Statistics with SAS Applications

BOOTSTRAP 163

outs=bootstrap_replicates;

var hours score;

by replicate;

run;

data bootstrap_replicates;

set bootstrap_replicates;

if(_type_=’CORR’ and _name_=’hours’);

spearman_corr=score;

keep spearman_corr;

run;

proc univariate data=bootstrap_replicates;

var spearman_corr;

output out=CI pctlpre=CI_99_ pctlpts=0.5 99.5

pctlname=lower upper;

run;

proc print data=CI;

run;

The output is

CI_99_ CI_99_

lower upper

-0.71292 0.81942

Note that the jackknife method gives a wider 99% confidence interval. 

Exercises for Chapter 8

Exercise 8.1 For the data in Exercise 7.1, give a 95% confidence interval for the

mean time between eruptions of Old Faithful Geyser, using the jackknife estimation

procedure.

Exercise 8.2 Use the data in Exercise 7.2 to construct a 99% confidence interval ...