SUBJECT:  Bootstrasp programming
AUTHOR:   Lawrence C. Hamilton
	  Department of Sociology
	  University of New Hampshire
SUPPORT:  none
 
COMMENTS:
Create a new directory, name of your choosing, and copy the materials stored
here into it.  Use this new directory has your current directory when using
Stata and these materials.

Programs example1.ado-example5.ado, and the data files source.dta and 
boot4.dta, are explained in more detail in "Bootstrap Programming" (STB-4, 
November 1991).  Example1.ado-example5.ado are problem-specific programs, but
they could easily be modified with a text editor to change the number of 
iterations, calculation details, data source, or the model being tested.

example1.ado
     Bootstraps the mean of a variable named X, from a dataset called 
     source.dta, B=1,000 times.  Results are saved in boot1.dta.

example2.ado
     Bootstraps a two-variable regression using data resampling.  Regresses Y 
     on X, from source.dta, B=1,000 times.  Results are saved in boot2.dta.

example3.ado
     Bootstraps two-variable regression using residual resampling, storing 
     results in boot3.dta.

example4.ado
     Bootstrap within Monte Carlo:  generates artificial samples of n=80 cases
     each, according to model [7].  Performs data resampling B=2,000 times on
     each sample, and records results to evaluate how often standard or boot-
     strap confidence intervals truly contain the regression parameter b=3.
     Boot4.dta, included with this disk, contains results from 400 Monte Carlo 
    samples.

     Warning:  as written, this program may run for hours or days, depending 
     on your computer's speed.  It needs at least 3.2 megabytes of disk space.

example5.ado
     Monte Carlo simulation of the sampling distribution of OLS estimates of 
     the slope in model [7].  Saves results in boot5.dta.

boot4.dta
     Results from example4.ado:  bootstrapping 400 Monte Carlo samples of 
     model [7], and calculating width and coverage of the resulting confidence
     intervals.

source.dta
     Data on income, density, and air pollution levels in 21 New York 
     counties, adapted from:

           Zupan, Jeffrey M. (1973).  The Distribution of Air
           Quality in the New York Region.  Baltimore:  Johns
           Hopkins University Press.