whatsnew8to9.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

	    see {help screeplot:{bf:[MV] screeplot}}.

{p 9 13 2}
	b.  Options {cmd:level()}, {cmd:blanks()}, {cmd:novce}, and 
            {cmd:norotated} allow more flexible control of the displayed 
            results.

{p 9 13 2}
	c.  Option {opt components(#)} 
            specifies the number of components to retain
            and is a synonym for old option {cmd:factor()}.

{p 9 13 2}
	d.  Options {cmd:tol()} and {cmd:ignore} provide advanced control 
            for computationally difficult problems.

{p 9 9 2}
     See {help pca:{bf:[MV] pca}}
     for more information.

{p 4 9 2} 
17.  New {cmd:estat} commands for use after {cmd:pca} and {cmd:pcamat} provide
     additional statistics and results:

{p 9 13 2}
    a.  {cmd:estat} {cmd:loadings} reports the component loading matrix in
        any of several available normalizations of the columns (eigenvectors).

{p 9 13 2}
    b.  {cmd:estat} {cmd:rotatecompare} reports the unrotated (principal)
        components next to the most recent rotated components.

{p 9 9 2}
     See {help pca postestimation:{bf:[MV] pca postestimation}}
     for more information.

{p 4 9 2} 
18.  New {cmd:estat} commands for use after any factor analysis or 
     any principal components analysis (that is, after 
     {cmd:factor} or {cmd:factormat} or after {cmd:pca} or {cmd:pcamat})
     provide
     additional statistics and results:

{p 9 13 2}
    a.  {cmd:estat} {cmd:anti} reports the anti-image correlation and
        anti-image covariance matrices.

{p 9 13 2}
    b.  {cmd:estat} {cmd:kmo} reports the Kaiser-Meyer-Olkin measure of
        sampling adequacy.

{p 9 13 2}
    c.  {cmd:estat} {cmd:residuals} reports the difference between the
        observed correlation or covariance matrix and the fitted (reproduced)
        matrix using the retained factors.

{p 9 13 2}
    d.  {cmd:estat} {cmd:smc} reports the squared multiple correlations (SMC)
        between each variable and all other variables.  SMC is a theoretical
        lower bound for communality, so it is an upper bound for the
        unexplained variance.

{p 9 9 2}
     See 
     {help factor postestimation:{bf:[MV] factor postestimation}} and 
     {help pca postestimation:{bf:[MV] pca postestimation}}
     for more information.

{p 4 9 2} 
19.  Three new graphs are available after any factor analysis ({cmd:factor}
     and {cmd:factormat}) or after any principal components analysis
     ({cmd:pca} and {cmd:pcamat}):

{p 9 13 2}
    a.  {cmd:scoreplot} graphs scatterplots comparing each pair of factors or
        components; see {help scoreplot:{bf:[MV] scoreplot}}.

{p 9 13 2}
    b.  {cmd:loadingplot} graphs scatterplots comparing loadings for each pair
        of factors or components; see {help scoreplot:{bf:[MV] scoreplot}}.

{p 9 13 2}
    c.  {cmd:screeplot} plots the eigenvalues of a covariance or
        correlation matrix; see {help screeplot:{bf:[MV] screeplot}}.
	({cmd:screeplot} replaces {cmd:greigen} and has more features;
	{cmd:greigen} continues to work but is undocumented.)

{p 4 9 2} 
20.  New command {cmd:rotate} performs orthogonal and oblique rotations
     after {helpb factor}, {helpb factormat}, {helpb pca}, and {helpb pcamat}.
     Available rotations include varimax, quartimax, equamax, parsimax,
     minimum entropy, Comrey's tandem 1 and 2, promax power, biquartimax,
     biquartimin, covarimin, oblimin, factor parsimony, Crawford-Ferguson
     family, Bentler's invariant pattern, oblimax, quartimin, and target and
     partial-target matrices; see {help rotate:{bf:[MV] rotate}}.

{p 9 9 2} 
    New command {cmd:rotatemat} performs these same linear
    transformations (rotations) on any Stata matrix.


{marker survival}{...}
{title:What's new:  Survival analysis}

{p 5 9 2}
1.  The {cmd:[ST]} manual now has a glossary that defines commonly used terms
    in survival (or duration) analysis and often explains how these terms are
    used in the manual; see the glossary of {bf:[ST]}.

{p 5 9 2}
2.  New command {cmd:estat} can be used after {cmd:stcox} and {cmd:streg}.
    In addition to the standard {cmd:estat} statistics -- information
    criteria, estimation sample summary, and formatted variance-covariance
    matrix (VCE) -- statistics specific to the proportional hazards estimator
    are available after {cmd:stcox}.  These include

{p 9 13 2}
	a.  {cmd:estat concordance} computes 
            Harrell's C and Somer's D statistics measuring concordance -- 
	    agreement of predictions with observed failure order.

{p 9 13 2}
        b.  {cmd:estat phtest} replaces the existing
            {cmd:stphtest} for computing tests and graphs of the proportional
            hazards assumption.  {cmd:stphtest} continues to work.

{p 9 9 2}
        See
        {helpb stcox postestimation:[ST] stcox postestimation}
        and 
        {helpb streg postestimation:[ST] streg postestimation}.

{p 5 9 2}
3.  Existing command {cmd:sts graph} has new options {cmd:cihazard} 
    and {cmd:per(}{it:#}{cmd:)}.  {cmd:cihazard} draws pointwise confidence
    bands around the smoothed hazard function, and {cmd:per()} specifies the
    units used to report the survival or failure rate.  See 
    {helpb sts:[ST] sts}.

{p 5 9 2}
4.  Existing command {cmd:stcurve} now
    plots over an evenly spaced grid, producing smooth curves, even in
    small samples;
    see {helpb stcurve:[ST] stcurve}.

{p 5 9 2}
5.  Existing command {cmd:sts graph} has new options {cmd:atriskopts()} and
    {cmd:lostopts()} that let you control how the labels for at-risk and lost
    observations look (their color, font size, etc.); see 
    {helpb sts:[ST] sts}.

{p 5 9 2}
6.  Existing command {cmd:stci} has new options for controlling how the
    plotted survival line looks (color, thickness, etc.) and for adding
    titles, controlling legends, and all other characteristics of the graph;
    see {helpb stci:[ST] stci}.


{marker general}{...}
{title:What's new:  General-purpose statistics}

{p 5 9 2}
1.  New estimation command {cmd:asmprobit} fits multinomial probit (MNP)
    models to
    categorical data and is frequently used in choice-based modeling.
    {cmd:asmprobit}
    allows several correlation structures for the alternatives,
    including completely unstructured, where all possible
    correlations are estimated.  It also allows for either heteroskedastic or
    homoskedastic variances among the alternatives and allows arbitrary
    patterns within the alternative variances or correlations.
    {cmd:asmprobit}'s syntax makes specifying both case-specific
    and alternative-in-case-specific regressors easy.

{p 9 9 2}
    In addition to common postestimation commands, such as {cmd:mfx} 
    for
    computing marginal effects, new command {cmd:estat} 
    provides
    additional statistics and results:

{p 9 13 2}
    a.  {cmd:estat} {cmd:alternatives} reports summary statistics about each
	of the the alternatives and provides a mapping between the
	index numbers labeling the alternatives and their associated values
	and labels in the dataset.

{p 9 13 2}
    b.  {cmd:estat} {cmd:covariance} computes and reports the estimated
        covariance matrix for the alternatives.

{p 9 13 2}
    c.  {cmd:estat} {cmd:correlation} reports the correlations among the
        alternatives in matrix form.

{p 9 9 2}
    Predicted statistics after {cmd:asmprobit} include the linear predictor,
    the probability an alternative is selected, and the standard error of the
    linear predictor.

{p 9 9 2}
    See 
    {help asmprobit:{bf:[R] asmprobit}},
    and 
    {help asmprobit postestimation:{bf:[R] asmprobit postestimation}}.

{p 5 9 2}
2.  New estimation command {cmd:mprobit} also fits multinomial probit models
    to categorical data but in the simplified situation of having only
    case-specific covariates (as with the multinomial logistic regression,
    {cmd:mlogit}).
    Maximizing the likelihood is much faster in such cases
    because the numeric approximation to the likelihood is simpler.
    See 
    {help mprobit:{bf:[R] mprobit}}.

{p 5 9 2}
3.  New estimation command {cmd:slogit} fits the stereotype logistic regression
    model for categorical dependent variables.  This model can be viewed as
    either a generalization of the multinomial logistic regression model
    ({cmd:mlogit}) or a generalization of the ordered logistic regression
    model ({cmd:ologit}) that relaxes the proportional-odds assumption. 
    See {help slogit:{bf:[R] slogit}}.

{p 9 9 2}
    Predicted statistics after {cmd:slogit} include the linear predictor,
    the probability of any or all outcomes, and the standard error of the
    linear predictor.
    See {help slogit postestimation:{bf:[R] slogit postestimation}}.

{p 5 9 2}
4.  New estimation command {cmd:ivprobit} fits probit regression models of
    binary outcomes with endogenous regressors.  Estimation can be performed
    by maximum likelihood estimation (MLE) or by Newey's minimum chi-squared
    two-step estimation, but some postestimation facilities,
    such as computing marginal effects with {cmd:mfx}, are available only
    after ML estimation -- the two-step estimator imposes a transformation
    that invalidates many postestimation results.  
    See {help ivprobit:{bf:[R] ivprobit}}.

{p 5 9 2}
5.  New estimation command {cmd:ivtobit} fits linear regression models with
    censored dependent variables by maximum likelihood estimation or by
    Newey's minimum chi-squared two-step estimation (but see the note about the
    the two-step estimator in 4 above).  
    See {help ivtobit:{bf:[R] ivtobit}}.

{p 5 9 2}
6.  New estimation command {cmd:ztp} fits a zero-truncated Poisson
    model of event counts with truncation at zero.

{p 9 9 2}
    Predicted statistics after {cmd:ztp} include the linear predictor and its
    standard error, the predicted number of events, the incidence rate, the
    conditional mean, and the likelihood score
    See {help ztp:{bf:[R] ztp}}
    and 
    {help ztp postestimation:{bf:[R] ztp postestimation}}.

{p 5 9 2}
7.  New estimation command {cmd:ztnb} fits a zero-truncated negative
    binomial model of event counts with truncation at zero and over or under
    dispersion.

{p 9 9 2}
    Predicted statistics after {cmd:ztnb} include the linear predictor and its
    standard error, the predicted number of events, the incidence rate, the
    conditional mean, and the likelihood scores
    See {help ztnb:{bf:[R] ztnb}}
    and
    {help ztnb postestimation:{bf:[R] ztnb postestimation}}.

{p 5 9 2}
8.  New estimation commands {cmd:mean}, {cmd:ratio}, {cmd:proportion}, and
    {cmd:total} estimate means, ratios, proportions, and totals over the
    entire sample or over groups within the sample.  When estimating over
    groups, the entire covariance matrix (VCE) is estimated.  These are 
    full estimation commands that support a range of postestimation facilities,
    such as linear and nonlinear tests among the groups ({helpb test} and
    {helpb testnl}) and linear and nonlinear combinations of group-level
    statistics ({helpb lincom} and {helpb nlcom}).  All four commands support
    several SE and VCE estimates:  robust, cluster-robust, bootstrap,
    jackknife, and observed information matrix (the default).  

{p 9 9 2}
    {cmd:mean}, {cmd:ratio}, and {cmd:proportion} also support direct
    standardization across strata (groups) using the {cmd:stdize()} and
    {cmd:stdweight()} options.

{p 9 9 2}
    See {help mean:{bf:[R] mean}},
        {help ratio:{bf:[R] ratio}},
        {help proportion:{bf:[R] proportion}},
    and {help total:{bf:[R] total}}.

{p 5 9 2}
9.  To avoid conflict with the new {cmd:mean} command, existing command
    {cmd:means} has been renamed {cmd:ameans}, with synonyms
    {cmd:gmeans} and {cmd:hmeans}.

{p 4 9 2}
10.  Existing command {cmd:nl} has a new syntax that makes estimating
    nonlinear least-squares regressions easier.  For most models, estimation
    is now as easy as typing the nonlinear expression.  Full programmability
    has been retained for complex models, and the old syntax continues to work.

{p 9 9 2}
    {cmd:nl} also now supports robust (Huber/white/sandwich) and
    cluster-robust SE and VCE estimates, including two popular 
    adjustments that can dramatically improve the small-sample
    performance of robust SE and VCE estimates.  
    
{p 9 9 2}
    A number of new reporting and estimation options have also been added.
    See {help nl:{bf:[R] nl}}.


{p 4 9 2}
11.  New option {cmd:vce()} selects how standard errors (SEs) and covariance
     matrix of the estimated parameters are estimated by most estimation
     commands.
     Choices are {cmd:vce(oim)}, 
     {cmd:vce(opg)}, 
     {cmd:vce(robust)}, 
     {cmd:vce(jackknife)}, and 
     {cmd:vce(bootstrap)}, although the choices can vary estimator by 
     estimator.
     {cmd:vce(robust)} is a synonym for {cmd:robust}, and you can use either.
     What is new are {cmd:vce(jackknife)} and {cmd:vce(bootstrap)}.