The name is absent



24


Stata Technical Bulletin


STB-4


display (_b [XX] -7._orb) / (_b [XX] ∕sqrt (.result (6) ) )

* Previous four lines display the original-sample regression slope

* (.orb), original-sample standard error (.orSE), bootstrap

* regression slope, and bootstrap studentized slope,
log off

macro define .bsample=%.bsample+l
>

log close
drop .all
infile orb orSE bootb stub using bstemp.log
save mcit%.mcit, replace

* Each Monte Carlo iteration saves a dataset containing

* results from 2,000 resamplings,
macro define .mcit=%.mcit+l

>

macro define .mcit=l

capture erase mctemp.log

log using mctemp.log

log off

while %.mcit<101 -(

quietly use mcit%.mcit, clear
log on

display %.mcit

quietly summ orb

display .result(3)

quietly summ orSE

display .result(3)

quietly summ bootb, detail

display .result(3)

display sqrt(.result(4))

display .result(7)

display .result(13)

quietly summ stub, detail

display .result(7)

display .result(13)

display
log off

* Preceding lines display the Monte Carlo sample number, original-sample

* slope and standard error, bootstrap mean and standard deviation of

* regression slope; and bootstrap percentiles and studentized percentiles,
macro define .mcit=%.mcit+l

>

log close

drop .all

infile iterate orb orSE bootb bootSE p05 p95 t05 t95 ∕*

*/ using mctemp.log;

label variable orb "original sample b coefficient"

label variable orSE "original sample SE of b"

label variable bootb "mean bootstrap b coefficient"

label variable bootSE "SD of bootstrap b coefficient"

label define correct 0 "wrong" 1 "correct"
*

generate stanlo=orb-l.665*orSE

generate stanhi=orb+l.665*orSE

generate Stanwide=Stanhi-Stanlo

label variable Stanwide "width standard t interval"

generate Stancor=O

replace Stancor=I if stanlo<3 & 3<stanhi

label variable Stancor "standard 90% c.i. correct?"
label values Stancor correct

*

generate perwide=p95-p05

label variable perwide "width 5th-95th percentile"

generate percor=0

replace percor=l if p05<3 & 3<p95

label variable percor "percentile 90% c.i. correct?"
label values percor correct

*

generate hyblo=orb-(p95-orb)

generate hybhi=orb-(p05-orb)

generate hybwide=hybhi-hyblo

label variable hybwide "width hybrid percentile interval"

generate hybcor=0

replace hybcor=l if hyblo<3 & 3<hybhi



More intriguing information

1. The name is absent
2. Private tutoring at transition points in the English education system: its nature, extent and purpose
3. The name is absent
4. A Consistent Nonparametric Test for Causality in Quantile
5. ASSESSMENT OF MARKET RISK IN HOG PRODUCTION USING VALUE-AT-RISK AND EXTREME VALUE THEORY
6. The name is absent
7. MICROWORLDS BASED ON LINEAR EQUATION SYSTEMS: A NEW APPROACH TO COMPLEX PROBLEM SOLVING AND EXPERIMENTAL RESULTS
8. The name is absent
9. The name is absent
10. Economic Evaluation of Positron Emission Tomography (PET) in Non Small Cell Lung Cancer (NSCLC), CHERE Working Paper 2007/6