whatsnew7to8.hlp

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

7.  Existing command {cmd:jknife} now accepts the {cmd:cluster()} option; see
    help {help jknife}.

{p 6 10 2}
8.  New command {cmd:permute} estimates p-values for permutation tests based
    on Monte Carlo simulations.  These estimates can be one sided or two
    sided.  See help {help permute}.

{p 6 10 2}
9.  Existing command {cmd:sample} has new option {cmd:count} that allows
    samples of the specified number of observations (rather than a percentage)
    to be drawn.  In addition, {cmd:sample} now allows the
    {bind:{cmd:by} {it:varlist}{cmd::}} prefix as an alternative to the
    already existing {cmd:by(}{it:varlist}{cmd:)} option; both do the same
    thing.  See help {help sample}.

{p 5 10 2}
10.  New command {cmd:simulate} replaces {cmd:simul} and provides improved
     syntax for specifying simulations; see help {help simulate}.

{p 5 10 2}
11.  Existing command {cmd:statsby} has a new syntax, new options, and now
     allows time-series operators; see help {help statsby}.

{p 5 10 2}
12.  The new {cmd:estimates} command provides a new, consistent way to store
     and refer to estimation results.  Post-estimation commands that make
     comparisons across models, such as {cmd:lrtest} and {cmd:hausman},
     previously had their own idiosyncratic ways to store and refer to
     estimation results.  These commands now support a unified way of
     retrieving estimation results utilizing the new {cmd:estimates} suite.

{p 10 10 2}
     Under the new scheme, after fitting a model, you can type

{p 14 18 2}
	{cmd:. estimates store} {it:name}

{p 10 10 2}
     to save the results.  At some point later in the session, you can type

{p 14 18 2}
	{cmd:. estimates restore} {it:name}

{p 10 10 2}
     to get back the estimates.  You can redisplay estimates (without
     restoring them) by typing

{p 14 18 2}
	{cmd:. estimates replay} {it:name}

{p 10 10 2}
     Other estimation manipulation commands are provided; see help
     {help estimates}.

{p 10 14 2}
     a.  Existing command {help lrtest} has been modified to have syntax

{p 18 22 2}
	{cmd:lrtest} {it:name} {it:name}

{p 10 14 2}
     b.  Existing command {help hausman} has been modified to have syntax

{p 18 22 2}
	{cmd:hausman} {it:name} {it:name}

{p 10 14 2}
     c.  The new {cmd:estimates for} command can be used in front of any
	 post-estimation command, such as {help test} or {help predict}, to
	 perform the action on the specified set of estimation results,
	 without disturbing the current estimation results.  With
	 {cmd:estimates for}, you can type such things as

{p 18 22 2}
	{cmd:. estimates for} {it:earlierresults}{cmd:: predict expected}

{p 14 14 2}
     See help {help estimates}.

{p 10 14 2}
     d.  The new {cmd:estimates stats} command displays the Akaike Information
	 Criterion (AIC) and Schwarz Information Criterion (BIC) model
	 selection indexes.  See help {help estimates}.

{p 5 10 2}
13.  Existing command {cmd:lrtest} now supports composite models specified by
     a parenthesized list of model names.  In a composite model, it is assumed
     that the log likelihood and dimension of the full model are obtained as
     the sum of the log likelihoods and the sum of the dimensions of the
     constituent models.

{p 10 10 2}
     {cmd:lrtest} has a new {cmd:stats} option to display statistical
     information about the unrestricted and restricted models, including the
     AIC and BIC model selection statistics.  See help {help lrtest}.

{p 5 10 2}
14.  {cmd:test} has improved syntax:

{p 10 14 2}
     a.  You may now type

{p 18 22 2}
	    {cmd:. test} {it:a} {cmd:=} {it:b}

{p 14 14 2}
	 for expressions {it:a} and {it:b}, or you may type

{p 18 22 2}
	    {cmd:. test} {it:a} {cmd:==} {it:b}

{p 14 14 2}
	 The use of {cmd:==} is more consistent with Stata's syntax that
	 treats {cmd:==} as indicating comparison and {cmd:=} as meaning
	 assignment.

{p 10 14 2}
     b.  You may now specify multiple tests on one line:

{p 18 22 2}
	 {cmd:. test} {cmd:(}{it:a} {cmd:==} {it:b} {cmd:==} {it:c}{cmd:)}

{p 18 22 2}
	 {cmd:. test} {cmd:(}{it:a} {cmd:==} {it:b}{cmd:)}
			{cmd:(}{it:c} {cmd:==} {it:d}{cmd:)}

{p 10 14 2}
     c.  {cmd:test} has new option {cmd:coef}, which specifies that the
	 constrained coefficients are to be displayed.

{p 10 14 2}
     d.  {cmd:test} has two new options for use with the
	 {cmd:test} {cmd:[}{it:eq1}{cmd:==}{it:eq2}{cmd:]} syntax:
	 {cmd:constant} and {cmd:common}.  {cmd:constant} specifies that
	 {cmd:_cons} should be included in the list of coefficients to be
	 tested.  {cmd:common} specifies that {cmd:test} restrict itself to
	 the coefficient in common between {it:eq1} and {it:eq2}.

{p 10 14 2}
     e.  {cmd:test} may now be used after survey estimation.

{p 10 14 2}
     f.  {cmd:test} has a new programmer's option
	 {cmd:matvlc(}{it:matname}{cmd:)}, which saves the variance-covariance
	 matrix of the linear combination(s).

{p 10 10 2}
     See help {help test}.

{p 5 10 2}
15.  {cmd:testnl} now allows typing
     {cmd:testnl} {it:exp}{cmd:==} {it:exp} {cmd:==} ... {cmd:==} {it:exp}
     to test whether two or more expressions are equal.  Single equal signs
     may be used:
     {cmd:testnl} {it:exp}{cmd:=} {it:exp} {cmd:=} ... {cmd:=} {it:exp}.

{p 10 10 2}
     In addition, {cmd:testnl} has new option {cmd:iterate(}{it:#}{cmd:)} for
     specifying the maximum number of iterations used to find the optimal step
     size in the calculation of the numerical derivatives of the expressions
     to be tested.  See help {help testnl}.

{p 5 10 2}
16.  {cmd:testparm} has new option {cmd:equation()} for use after fitting
     multiple-equation models such as {help mvreg}, {help mlogit},
     {help heckman}, etc.  It specifies the equation for which the all-zero or
     all-equal hypothesis is to be tested.  See help {help test}.

{p 5 10 2}
17.  {cmd:lincom} now works after {help anova} and after all survey
     estimators; see help {help lincom}.

{p 5 10 2}
18.  {cmd:bitest}, {cmd:prtest}, {cmd:ttest}, and {cmd:sdtest} now allow
     {cmd:==} to be used wherever {cmd:=} is allowed in their syntax; See help
     {help bitest}, {help prtest}, {help ttest}, and {help sdtest}.

{p 5 10 2}
19.  New command {cmd:suest} is a post-estimation command that combines
     multiple estimation results (parameter vectors and their
     variance-covariance matrices) into simultaneous results with a single
     stacked parameter vector and a robust (sandwich) variance-covariance
     matrix. The estimation results to be combined may be based on different,
     overlapping, or even the same data.  After creating the simultaneous
     estimation results, one can use {help test} or {help testnl} to obtain
     Hausman-type tests for cross-model hypotheses.  {cmd:suest} supports
     survey data.  See help {help suest}.

{p 5 10 2}
20.  New command {cmd:imtest} performs the information matrix test for an a
     regression model.  In addition, it provides the Cameron-Trevedi
     decomposition of the IM-test in tests for heteroskedasticity, skewness,
     and kurtosis, and White's original heteroskedasticity test.  See help
     {help imtest}.

{p 5 10 2}
21.  New command {cmd:szroeter} performs Szroeter's test for
     heteroskedasticity in a regression model; see help {help szroeter}.

{p 5 10 2}
22.  Existing command {cmd:hettest} now provides option {cmd:rhs} to test for
     heteroskedasticity in the independent variables.  It now also supports
     multiple comparison testing.  See help {help hettest}.

{p 5 10 2}
23.  Existing command {cmd:tabulate} has output changes, new features, and
     expanded limits.

{p 10 14 2}
     a.  Three new statistics are available for twoway tabulations:
	 {cmd:expected}, {cmd:cchi2}, and {cmd:clrchi2}.  {cmd:expected}
	 reports the expected number in each cell.  {cmd:cchi2} reports the
	 contribution to Pearson's chi-squared.  {cmd:clrchi2} reports the
	 contribution to the likelihood-ratio chi-squared.

{p 10 14 2}
     b.  New options {cmd:key} and {cmd:nokey} force or suppress a key
	 explaining the entries in the table.

{p 10 14 2}
     c.  Twoway tabulations now respect {cmd:set linesize}, meaning you can
	 produce wide tables.

{p 10 14 2}
     d.  Both oneway and twoway tabulations now put commas in the reported
	 frequency counts.

{p 10 14 2}
     e.  {cmd:tabulate} for oneway tabulations has new option {cmd:sort},
	 which puts the table in descending order of frequency.

{p 10 14 2}
     f.  {cmd:tabulate} has expanded limits:

		  {c TLC}{hline 19}{c TT}{hline 8}{c TT}{hline 13}{c TRC}
		  {c |} Flavor            {c |}  1-way {c |}    2-way    {c |}
		  {c LT}{hline 19}{c +}{hline 8}{c +}{hline 13}{c RT}
		  {c |} {help SpecialEdition:Stata/SE}          {c |} 12,000 {c |} 12,000 x 80 {c |}
		  {c |} Intercooled Stata {c |}  3,000 {c |}    300 x 20 {c |}
		  {c |} Small Stata       {c |}    500 {c |}    160 x 20 {c |}
		  {c BLC}{hline 19}{c BT}{hline 8}{c BT}{hline 13}{c BRC}

{p 10 10 2}
     See help {help tabulate}.

{p 5 10 2}
24.  Existing command {cmd:tabstat} has new options {cmd:statistics(variance)}
     and {cmd:statistics(semean)} which display the variance and the standard
     error of the mean.  (Also provided is new option
     {cmd:varwidth(}{it:#}{cmd:)}, specifying the number of characters used to
     display variable names.)  See help {help tabstat}.

{p 5 10 2}
25.  Existing command {cmd:roctab} has new option {cmd:specificity} to graph
     sensitivity versus specificity, instead of the default sensitivity versus
     (1-specificity); see help {help roctab}.

{p 5 10 2}
26.  Existing command {cmd:ologit} now has option {cmd:or} to display results
     as odds ratios (display exponentiated coefficients); see help
     {help ologit}.

{p 5 10 2}
27.  New command {cmd:lowess} replaces old command {cmd:ksm}.  {cmd:lowess}
     allows {cmd:graph} {cmd:twoway}'s {cmd:by()} option and is much faster
     than {cmd:ksm}; see help {help lowess}.

{p 5 10 2}
28.  Existing command {cmd:kdensity} has been rewritten so that it executes
     faster; see help {help kdensity}.

{p 5 10 2}
29.  Existing command {cmd:intreg} now applies constraints in the same manner
     as all other estimation commands, and existing command {cmd:mlogit} now
     allows constraints with constants; see help {help intreg} and
     {help mlogit}.

{p 5 10 2}
30.  New command {cmd:pca} performs principal components analysis, replacing
    {cmd:factor,} {cmd:pc}; see help {help pca}.

{p 5 10 2}
31.  Existing command {cmd:ml} {cmd:maximize} and all estimators using {cmd:ml}
     have a new tolerance option {cmd:nrtolerance(}{it:#}{cmd:)} for
     determining convergence.  Convergence is declared when
     {bf:g}*inv({bf:H})*{bf:g}' < {cmd:nrtolerance(}{it:#}{cmd:)}, where {bf:g}
     represents the gradient vector and {bf:H} the Hessian matrix; see help
     {help maximize}.

{p 5 10 2}
32.  Existing command {cmd:mfx} will now use {cmd:pweight}s or {cmd:iweight}s
     when calculating the means or medians for the {it:atlist} following an
     estimation command that used {cmd:pweight}s or {cmd:iweight}s.
     Previously, only {cmd:fweight}s and {cmd:aweight}s were supported.  See
     help {help mfx}.

{p 5 10 2}
33.  Existing command {cmd:adjust} now allows the {cmd:pr} option to display
     predicted probabilities when used after {help svylogit},
     {help svyprobit}, {help xtlogit}, and {help xtprobit}.  See help
     {help adjust}.

{p 5 10 2}
34.  The existing regression diagnostics commands {help acprplot},
     {help cprplot}, {help hettest}, {help lvr2plot}, {help ovtest},
     {help rvfplot}, and {help rvpplot} have been extended to work after
     {help anova}.  In addition, {help cprplot} and {help acprplot} have new
     options {cmd:lowess} and {cmd:mspline} that allow putting a lowess curve
     or median spline through the data.  See help {help regdiag}.

{p 5 10 2}
35.  Existing command {cmd:ranksum} has new option {cmd:porder} that estimates
     P(x_1>x_2); see help {help signrank}.

{p 5 10 2}
36.  Existing command {cmd:poisgof} has new option {cmd:pearson} to request
     the Pearson chi-squared goodness-of-fit statistic; see help {help poisson}.

{p 5 10 2}
37.  Existing command {cmd:binreg} now respects the {cmd:init()} option; see
     help {help binreg}.

{p 5 10 2}
38.  Existing command {cmd:boxcox} now accepts {cmd:iweight}s; see help
     {help boxcox}.

{p 5 10 2}
39.  Existing commands {cmd:zip} and {cmd:zinb} now accept the
     {it:maximize_option} {cmd:from()} to provide starting values; see help
     {help zip}.

{p 5 10 2}
40.  Existing command {cmd:cnsreg} now accepts the {cmd:noconstant} option;
     see help {help cnsreg}.

{p 5 10 2}
41.  Existing command {cmd:hotel} has been renamed {cmd:hotelling}; {cmd:hotel}
    is now an abbreviation for {cmd:hotelling}; see help {help hotelling}.

{p 5 10 2}
42.  The {cmd:score()} option is now unified across all estimation commands.
     You must specify the correct number of score variables, and, in
     multiple-equation estimators, you may specify {it:stub}{cmd:*} to mean
     create new variables named {it:stub}{cmd:1}, {it:stub}{cmd:2}, ...

{p 10 10 2}
     Estimation commands now save in {cmd:e(scorevars)} the names of the score
     variables if {cmd:score()} was specified.

{p 5 10 2}
43.  Existing command {cmd:summarize} without the {cmd:detail} option now
     allows {cmd:iweight}s; see help {help summarize}.