|
|
|
||||
|
|
|||||
* Several commands have been eliminated and made into a prefix for other commands
* One example of this is the svy commands, which we will get to in a few minutes.
* Prefixes can now have options
* list of prefix commands
/*
by run command on subsets of data
statsby same as by, but collect statistics from each run
rolling run command on moving subsets, and collect statistics
bootstrap run command on bootstrap samples
jackknife run command on jackknife subsets of data
permute run command on random permutations
simulate run command on manufactured data
svy run command and adjust results for survey sampling
stepwise run command with stepwise variable inclusion/exclusion
xi run command after expanding factor variables & interactions
capture run command and capture its return code
noisily run command and show the output
quietly run command and suppress the output
version run command under specified version
*/
use http://www.ats.ucla.edu/stat/stata/notes/hsb2, clear * same as Stata 8 xi: regress write i.ses * naming the prefix xi , prefix(dum): regress write i.ses * making all of the dummies xi , noomit prefix(ndum): regress write i.ses, nocons * bootstrap and jackknife are now prefixes regress write read female bootstrap, reps(100) : regress write read female jackknife: regress write read female * the same commands with some options included so that the outputs are close together * and can be easily compared regress write read female, nohead bootstrap, reps(100) nohead nodots: regress write read female jackknife, nohead nodots: regress write read female mean read write math bootstrap, reps(100) nohead nodots: mean read write math jackknife, nohead nodots: mean read write math * this example shows getting a median (we need to do it this way because of our data set) bootstrap r(p50), reps(500) nodots: summarize write, detail * permute is now a prefix permute female sum=r(sum), right nodrop nodots reps(1000): summarize read if ~female permute female r(p50), reps(1000) right nodrop nodots: summarize read if ~female, detail * rolling is a new prefix * Los Angeles seasonal rainfall 1878 to present use http://www.ats.ucla.edu/stat/stata/seminars/stata9/larain, clear * desc list in 1/30 twoway scatter laseason year, xlabel(1900(50)2000) summarize tsset year rolling, window(20): summarize laseason list in 1/30 twoway scatter mean start * version is now a prefix use http://www.ats.ucla.edu/stat/stata/seminars/svy_stata_intro/oscs1, clear version 8.2: svyregress api00 awards meals
* using two prefixes * Note that the order of the prefixes matters. xi: svy: regress api00 i.cname growth
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services