[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: compare groups using bootstrap

From   John Winer <[email protected]>
To   [email protected]
Subject   st: compare groups using bootstrap
Date   Fri, 29 Jan 2010 23:13:50 -0800 (PST)

hello statalisters,

I am fairly new to stata and still trying to get to grips with it. I recently encountered a problem with bootstrap difference in means test. 

The code that I am using is based on several messages on the statalist. For example the one from brian poi of stata corp I have 10 numbers of such two group comparisions and this is my code.

for i=1(1)10 {
bootstrap r(t), reps(1000) seed(5432) saving(bs_1,replace):ttest var  if gid==`i', by(group) unequal unpaired

* this will output the results in a table
*! version 1.0 a 25 jan 2010
capture program drop bootout
program define bootout
version 9
        matrix b = e(b)
        matrix se = e(se)
        local names : colnames(se)
        local df = e(N) - colsof(b)
        tempname t lb ub
        local i = 1
        foreach var of local names {
                scalar `t' = (b[1, `i'] / se[1, `i'])
                scalar `lb' = b[1, `i'] - se[1, `i']*invttail(`df', 0.025)
                scalar `ub' = b[1, `i'] + se[1, `i']*invttail(`df', 0.025)
                di "`var'" _col(15) %9.0g b[1,`i'] ///
                           _col(26) %9.0g se[1, `i'] ///
                           _col(38) %6.2f `t' ///
                           _col(46) %6.3f 2*ttail(`df', abs(`t')) ///
                           _col(56) %9.0g `lb' ///
                           _col(68) %9.0g `ub'
                local i = `i' + 1

Is this the right way to do this?
Also the above bootout prog assumes that the distribution of the bootstrapped t is normal. in otherwords the CI that i construct is based on a normal disribution.

How does one use the BCa or the percentile CI and work out a t statistic and p value for the difference in mean?



*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index