stata4.key

是一个经济学管理应用软件 很难找的 但是经济学学生又必须用到

.k reg-ressions st_rpool collinear-ity multicol-linearity
.x
	10/99	Why do ^stcox^ and ^cox^ sometimes produce missing
		standard errors?
		FAQS/stat/coxmiss.html

.e FAQ
.c faq
.t Tests comparing levels of a cat. variable after ^anova^ or ^regress^
.a K. Higbee
.k stat-istics test-s anova reg-ressions indica-tor var-iables dum-my dum-mies
.k cat-egorical factor-s cell mean-s model-s over-parameterized param-eterized
.k level-s anal-ysis of variance-s anocova ancova test-ing
.x
	10/99	How can I form various tests comparing the different
		levels of a categorical variable after ^anova^ or
		^regress^?
		FAQS/stat/test1.html

.e FAQ
.c faq
.t Fitting ordered logistic and probit models with constants
.a M. Inlow and R. Cong
.k stat-istics ologit oprobit order-ed logit logis-tic reg-ressions probit
.k constant-s
.x
	8/99	How can I convert Stata's parameterization of ordered
		probit and logistic models to one in which a constant
		is estimated?
		Why is there no constant term reported in ^ologit^ and
		^oprobit^?
		FAQS/stat/ologit_con.html

.e FAQ
.c faq
.t Comparing xtgls with regress, cluster()
.a V. Wiggins
.k stat-istics xt-gls clus-ters robust stand-ard err-ors std reg-ressions
.k hetero-scedasticity hetero-skedasticity panel-s correct-ed pcse fgls
.x
	8/99	How does ^xtgls^ differ from regression clustered with
		robust standard errors?
		FAQS/stat/xtgls_rob.html

.e FAQ
.c faq
.t Missing values reported by ^stsum^
.a M. Cleves
.k stat-istics stsum st surviv-al time miss-ing
.x
	7/99	Why does ^stsum^ sometimes report missing values for
		the percentiles of survival time?
		FAQS/stat/stsum.html

.e FAQ
.c faq
.t Relationship between the chi-squared and F distributions
.a W. Gould
.k stat-istics test-s chi-squared sq-uared chi-2 2 f-statistics distrib-utions
.k test-ing
.x
	7/99	Why does ^test^ sometimes produce chi-squared and other
		times F statistics?
		How are the chi-squared and F distributions related?
		FAQS/stat/wald.html

.e FAQ
.c faq
.t Chow and Wald tests
.a W. Gould
.k stat-istics chow test-s wald struc-tural change model stabil-ity robust
.k test-ing
.x
	7/99	How can I do a Chow test with the robust variance
		estimates, that is, after estimating with
		^regress, robust^?
		FAQS/stat/chow2.html

.e FAQ
.c faq
.t Mills' ratios and censoring direction in the Heckman selection model
.a V. Wiggins
.k stat-istics mill-s ratio inv-erse nonselec-tion hazard selec-tion heckman
.k samp-le censor-ing censor-ed direction model
.x
	5/99	Why are there so many formulas for the inverse of
		Mills' ratio?
		What if I have censoring from above/below in my
		Heckman selection model?
		FAQS/stat/invmills.html

.e FAQ
.c faq
.t ^xtreg^ with the ^mle^ option versus ^xtreg^ with the ^re^ option
.a V. Wiggins
.k stat-istics est-imation est-imated est-imates xt-reg ml-e re breusch pagan
.k cross section-al time-series series panel-s data-sets set-s longi-tudinal
.k reg-ressions hausm-ans robust meth-ods gls general-ized least sq-uares
.k specifica-tion test-s test-ing between within linear max-imum like-lihood
.k lagrange multipli-er mixed est-imators pred-ictions goodness fit
.k rand-om effect-s model-s pop-ulation ave-raged fix-ed between linear
.x
	4/99	Why does ^xtreg^ with the ^mle^ option produce different
		results from ^xtreg^ with only the ^re^ option?
		FAQS/stat/mle_vs_gmm.html

.e FAQ
.c faq
.t Interpreting standard errors produced by ^xtgee^ and SAS's GENMOD
.a J. Hardin
.k stat-istics est-imation est-imated est-imates sas xt-gee genmod gee
.k stand-ard err-ors std general-ized eq-uations
.k prob-ability weight-ing robust meth-ods samp-ling samp-les clus-ters
.k weight-ed pw-eights cross section-al time series panel-s data longi-tudinal
.x
	4/99	Why do Stata's ^xtgee^ standard errors differ from those
		reported by SAS's PROC GENMOD?
		FAQS/stat/xtgeesas.html

.e FAQ
.c faq
.t Prediction confidence intervals after ^logistic^ regression
.a M. Inlow
.k stat-istics est-imation est-imated est-imates logit prob-ability
.k conf-idence interv-als pred-ictions prob-abilities pred-icted reg-ressions
.k logis-tic
.x
	4/99	How do I obtain confidence intervals for the predicted
		probabilities after ^logistic^ regression?
		FAQS/stat/prep.html

.e FAQ
.c faq
.t Correlations between the factors after an oblique rotation
.a G. Fisher and K. Higbee
.k factor-s rotate stat-istics anal-ysis rotation-s load-ings corr-elations
.k multiv-ariate
.x
	4/99	How can I obtain the correlation between the factors
		after an oblique rotation?
		FAQS/stat/correlation.html

.e FAQ
.c faq
.t Clarification on analytic weights with linear regression
.a W. Gould
.k weight-s analytic reg-ressions linear stat-istics aweight-s est-imation
.k weight-ed
.x
	1/99	What is the effect of specifying ^aweight^s with ^regress^?
		FAQS/stat/crc36.html

.e FAQ
.c faq
.t Obtaining the variance-covariance matrix or coefficient vector
.a P. Lin
.k stat-istics est-imation est-imated est-imates variance-s cov-ariances
.k coef-ficients vec-tors matr-ix matr-ices param-eters
.x
	1/99	How can I get the variance-covariance matrix or
		coefficient vector?
		FAQS/stat/vc.html

.e FAQ
.c faq
.t Interpreting the cut points in ordered probit and logit
.a W. Gould
.k stat-istics est-imation est-imated est-imates probit logit order-ed
.k oprobit ologit logis-tic reg-ressions cut point-s
.x
	1/99	In ordered probit and logit, what are the cut points?
		FAQS/stat/oprobit.html

.e FAQ
.c faq
.t Stratified generalized Wilcoxon test
.a R. Thisted and M. Cleves
.k stat-istics surviv-al anal-ysis fail-ure time data strat-ified
.k wilcoxon test-s test-ing general-ized sas duration reliab-ility
.x
	1/99	Why do Stata and SAS differ in the results that they
		report for the stratified generalized Wilcoxon test
		for time-to-event data?
		FAQS/stat/wilcoxon.html

.e FAQ
.c faq
.t Failure time, censoring time, and entry time in the Cox model
.a W. Gould
.k stcox stat-istics est-imation est-imates
.k propor-tional haz-ards model-s st reg-ressions ratio-s rate-s
.k ml-e max-imum like-lihood epi-demiological epi-demiology
.k exp-onential distrib-utions weib-ull surviv-al fail-ure anal-ysis time
.k relativ-e risk-s graph-ics data kaplan meier duration
.k baseline vary-ing reg-ressors censor-ed censor-ing graph-s reliab-ility
.x
	1/99	Why can't a subject die at time 0?
		Why can't a subject enter and die at the same time?
		FAQS/stat/zerohelp.html

.e FAQ
.c faq
.t Comparing the p-values between a paired t test and a signrank
.a W. Sribney
.k signrank signtest rank-sums stat-istics test-s test-ing
.k equal-ity distrib-utions med-ians pop-ulations hyp-othesis hyp-otheses
.k match-ed pair-s mean-s sum-s nonpar-ametric order non param-etric
.k sign-edranks sign-ed p-values pair-ed stud-entst t-tests locat-ions
.x
	1/99	Is my boss correct in saying that the p-value given with
		a paired ^ttest^ should always be lower than the ^signrank^?
		FAQS/stat/signrank.html

.e FAQ
.c faq
.t Computing the Chow statistic
.a W. Gould
.k chow test-s test-ing stat-istics struc-tural change model stabil-ity
.x
	1/99	How can I compute the Chow test statistic?
		FAQS/stat/chow.html

.e FAQ
.c faq
.t Comparing random-effects and population-averaged models
.a W. Sribney
.k stat-istics est-imation est-imated est-imates model-s rand-om effect-s
.k pop-ulation ave-raged clus-ters re pa
.k cross section-al time-series series panel-s data longi-tudinal
.x
	1/99	What is the difference between random-effects and
		population-averaged estimators?
		FAQS/stat/repa.html

.e FAQ
.c faq
.t Confidence intervals for predicted probabilities after ^probit^
.a W. Sribney
.k stat-istics est-imation est-imated est-imates probit prob-ability
.k conf-idence interv-als pred-ictions prob-abilities pred-icted
.x
	1/99	How can I get confidence intervals for predicted
		probabilities after ^probit^?
		FAQS/stat/ciprob.html

.e FAQ
.c faq
.t Fitting gompit models with ^cloglog^
.a J. Hardin
.k stat-istics est-imation model-s gompit cloglog reg-ressions extreme
.k value complem-entary log-log bina-ry outcome-s dicho-tomous
.x
	1/99	How can I fit an extreme value or gompit regression?
		FAQS/stat/gompit.html

.e FAQ
.c faq
.t The Rasch model in Stata
.a J. Weesie
.k stat-istics rasch model-s bina-ry dicho-tomous succ-ess fail-ure
.k logit linear cond-itional fix-ed effect-s
.x
	1/99	Can Stata estimate a Rasch model?
		FAQS/stat/rasch.html

.e FAQ
.c faq
.t Interpreting ". . . completely determined" when running ^logistic^
.a W. Sribney
.k stat-istics est-imation est-imated est-imates logit logis-tic
.k reg-ressions model-s complete-ly determ-ined
.x
	1/99	What does completely determined mean in my ^logistic^
		regression output?
		FAQS/stat/logitcd.html

.e FAQ
.c faq
.t Comparison of std errors for robust, cluster, and standard estimators
.a W. Sribney
.k survey svy stat-istics weight-ed pw-eights clus-tering est-imators
.k prob-ability weight-ing robust meth-ods samp-ling samp-les clus-ters
.k huber white-s weight-ed stand-ard err-ors std bias-ed sandw-ich
.k variance-s cov-ariances
.x
	7/98	How can the standard errors with the ^cluster()^ option be
		smaller than those without the ^cluster()^ option?
		FAQS/stat/cluster.html

.e FAQ
.c faq
.t Multiple clustering levels
.a W. Sribney
.k survey svy stat-istics clus-ters clus-tering
.x
	5/98	Is it possible to analyze survey data with two or more
		levels of clustering with the svy commands?
		FAQS/stat/svy.html

.e FAQ
.c faq
.t Stepwise regression with the svy commands
.a W. Sribney
.k survey svy stat-istics step-wise reg-ressions model selec-tion
.x
	5/98	Is there a way in Stata to do stepwise regression with
		^svylogit^ or any of the svy commands?
		FAQS/stat/stepsvy.html

.e FAQ
.c faq
.t Problems with stepwise regression
.a W. Sribney
.k step-wise reg-ressions stat-istics model selec-tion
.x
	5/98	What are some of the problems with stepwise regression?
		FAQS/stat/stepwise.html

.e FAQ
.c faq
.t The svy command's handling of zero weights
.a W. Sribney
.k survey svy stat-istics zero weight-s samp-ling pw-eights prob-ability
.x
	4/98	Do the svy commands handle weights differently than
		non-svy commands do?
		FAQS/stat/zerowgt.html

.e FAQ
.c faq
.t Advantages of the robust variance estimator
.a W. Sribney
.k survey svy stat-istics robust est-imators variance-s
.k max-imum like-lihood est-imation est-imates ml-e logis-tic logit reg-ressions
.k huber white-s sandw-ich std stand-ard err-ors
.x
	1/98	What are the advantages of using the robust variance
		estimator over the standard maximum-likelihood variance
		estimator in logistic regression?
		FAQS/stat/robust_var.html

.e FAQ
.c faq
.t Maximum likelihood estimation
.a W. Sribney
.k survey svy stat-istics probit logit clus-ters clus-tering
.k max-imum like-lihood est-imation est-imates ml-e svyprobt svylogit
.x
	12/97	Are the estimates produced by ^probit^ and ^logit^ with
		the ^cluster()^ option true ml estimates?
		Is there a difference between the estimates produced by
		^svyprobt, psu()^ and ^probit, cluster()^?  (and, similarly,
		between ^svylogit, psu()^ and ^logit, cluster()^)?
		FAQS/stat/svyml.html

.e FAQ
.c faq
.t Interpreting the intercept in the fixed-effects model
.a W. Gould
.k stat-istics est-imation est-imated est-imates xt-reg fix-ed effect-s
.k model-s intercept fe
.k cross section-al time-series series panel-s data longi-tudinal
.x
	10/97	How can there be an intercept in the fixed-effects model
		estimated by ^xtreg, fe^?
		FAQS/stat/xtreg2.html

.e FAQ
.c faq
.t Likelihood-ratio test after survey/robust ML estimation
.a W. Sribney
.k ratio-s clus-tered clus-tering clus-ters survey svy prob-ability weight-ing
.k pw-eights est-imated est-imators robust meth-ods std err-ors stand-ard
.k test-s test-ing chi-squared sq-uared chi-2 2 wald samp-les model-s samp-ling
.k max-imum like-lihood stat-istics est-imation est-imates ml-e logit probit
.x
	8/97	Why should I not do a likelihood-ratio test after an ML
		estimation (e.g., ^logit^, ^probit^) with clustering or pweights?
		FAQS/stat/lrtest.html

.e FAQ
.c faq
.t Chi-squared test for models estimated with robust standard errors
.a W. Sribney
.k clus-tered clus-tering clus-ters survey svy prob-ability weight-ing ratio-s
.k pw-eights est-imated est-imators robust meth-ods std err-ors stand-ard
.k samp-les model-s samp-ling test-s test-ing chi-squared sq-uared chi-2 2 wald
.k max-imum like-lihood stat-istics est-imation est-imates ml-e logit probit
.x
	8/97	How do the ML estimation commands (e.g. ^logit^ and ^probit^)
		compute the model chi-squared test when they estimate robust
		standard errors on clustered data?
		FAQS/stat/chi2.html

.e FAQ
.c faq
.t Pseudo-R^^2 for ^tobit^
.a W. Sribney
.k stat-istics r-squared sq-uared pseudo-rsquared r-2 tobit est-imation
.x
	6/97	Why is the pseudo-R^^2 for ^tobit^ negative or greater
		than 1?
		FAQS/stat/pseudor2.html

.e FAQ
.c faq
.t The ^anova^ command and collinearity
.a W. Sribney
.k anova collinear-ity var-iables drop-ped stat-istics
.k anal-ysis variance-s est-imation est-imated est-imates multicol-linearity
.x
	3/97	How does the ^anova^ command handle collinearity?
		FAQS/stat/anova.html

.e FAQ
.c faq
.t Probability weights, analytic weights, and summary statistics
.a W. Sribney
.k prob-ability weight-s weight-ing analytic summ-ary stat-istics
.k summ-arize aweight-s pw-eights descrip-tive
.x
	3/97	Why doesn't ^summarize^ accept ^pweight^s?
		What does ^summarize^ calculate when you use ^aweight^s?
		FAQS/stat/supweight.html

.e FAQ
.c faq
.t Weighted estimation and ^xtgee^
.a J. Hardin
.k xt-gee xtcorr stat-istics est-imates est-imation
.k prob-ability weight-ing robust meth-ods samp-ling samp-les clus-ters
.k weight-ed pw-eights cross section-al time series panel-s data longi-tudinal
.k gee general-ized est-imation eq-uations variance-s est-imators scor-es
.k clus-tering clus-tered strat-ified strata stratum
.k huber white-s stand-ard err-ors std bias-ed sandw-ich
.k rand-om effect-s model-s
.k glm link func-tions distrib-utions family general-ized linear
.k gauss-ian logit probit gamma odds ratio-s incid-ence rate-s binom-ial