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

        Spearman correlations between the dissimilarities and the
        approximating distances for each object.

{p 9 13 2}
    c.  {cmd:estat} {cmd:pairwise} reports a set of statistics 
        for each pairwise comparison; it reports the dissimilarities, the
        approximating distances, and the raw residuals.

{p 9 13 2}
    d.  {cmd:estat} {cmd:quantiles} reports the quantiles of 
        the residuals for each observation (after {cmd:mds}) or object (after
        {cmd:mdslong} or {cmd:mdsmat}).

{p 9 13 2}
    e.  {cmd:estat} {cmd:stress} reports the Kruskal stress
        (loss) measure between the transformed dissimilarities and fitted
        distances per object.

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

{p 9 9 2}
    In addition, there are two new commands for graphing results from a
    multidimensional scaling:

{p 9 13 2}
    a.  {cmd:mdsconfig} plots the approximating Euclidean configuration of the
        first two dimensions; see 
	{help mds postestimation##mdsconfig:{bf:[MV] mds postestimation}}.

{p 9 13 2}
    b.  {cmd:mdsshepard} produces a Shepard diagram of the dissimilarities
        against the approximating Euclidean distances; see 
        {help mds postestimation##mdsshepard:{bf:[MV] mds postestimation}}.

{p 9 9 2}
    {cmd:predict} after any multidimensional-scaling command will
    produce 

{p 9 13 2}
    a.  variables containing the approximating configuration
	({cmd:predict} {it:newvarlist}{cmd:,} {cmd:config});

{p 9 13 2}
    b.  variables containing the dissimilarity, distance, and raw residuals
        ({cmd:predict} {it:newvarlist}{cmd:,} {cmd:pairwise})

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

{p 5 9 2}
2.  New commands {cmd:ca} and {cmd:camat} perform two-way correspondence
    analysis using any of several available forms of normalization.
    {cmd:ca} performs the analysis on the cross-tabulation of two categorical
    variables; {cmd:camat} performs the analysis on a matrix of
    counts; see {help ca:{bf:[MV] ca}} for more information on both.

{p 9 9 2}
    The following new {cmd:estat} commands work after {cmd:ca} and {cmd:camat}
    and provide additional statistics and results

{p 9 13 2}
    a.  {cmd:estat} {cmd:coordinates} reports the coordinates in both the row
        space and the column space.

{p 9 13 2}
    b.  {cmd:estat} {cmd:distances} reports the chi-squared distances between
        the row profiles and between the column profiles, including the
	distances to the marginal distributions (commonly called centers).
	Both observed or fitted profiles are available.

{p 9 13 2}
    c.  {cmd:estat} {cmd:inertia} reports the inertia contributions of the
        individual cells.

{p 9 13 2}
    d.  {cmd:estat} {cmd:profiles} reports the row profiles and column 
        profiles -- the conditional distributions, given the other dimension.

{p 9 13 2}
    e.  {cmd:estat} {cmd:summarize} reports summary information of the row
        and column variables over the estimation sample.

{p 9 13 2}
    f.  {cmd:estat} {cmd:table} reports the fitted correspondence table, 
    	the observed "correspondence" table, or the expected table under
	the assumption of independence.

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

{p 9 9 2}
    In addition, there are two new commands for graphing results from a
    correspondence analysis:

{p 9 13 2}
    a.  {cmd:cabiplot} produces a biplot of each row category and 
        each column category;
        see {help ca postestimation##cabiplot:{bf:[MV] ca postestimation}}.

{p 9 13 2}
    b.  {cmd:caprojection} produces a graph that shows the ordering of row
        categories 
        and column categories on each principal dimension of the analysis.
        Each principal dimension is represented by a vertical line; markers
        are plotted on the lines where the row categories and column 
        categories project
        onto the dimensions; see 
	{help ca postestimation##caprojection:{bf:[MV] ca postestimation}}.

{p 9 9 2}
    {cmd:predict} after {cmd:ca} and {cmd:camat} computes fitted values and
    row or column scores for any dimension; see
    {help ca postestimation##predict:{bf:[MV] ca postestimation}}.

{p 5 9 2}
3.  The new command {cmd:procrustes} performs Procrustean analysis for
    comparing and measuring the similarity between two sets of variables:
    source and target.  Two datasets can also be compared if the datasets
    are first merged by record.

{p 9 9 2}
    The following new {cmd:estat} commands work after {cmd:procrustes} and
    provide additional statistics and results:

{p 9 13 2}
    a.  {cmd:estat} {cmd:compare} reports fit statistics of the three
        transformations available in Procrustean analysis:  orthogonal,
	oblique, and unrestricted.

{p 9 13 2}
    b.  {cmd:estat} {cmd:mvreg} reports the multivariate regression that is
        related to the current Procrustean analysis.

{p 9 13 2}
    c.  {cmd:estat} {cmd:summarize} reports summary information of the two
        sets of variables over the estimation sample.

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

{p 9 9 2}
    New command {cmd:procoverlay} after {cmd:procrustes} creates an
    overlay graph comparing the target variables to the fitted values derived
    from the source variables; see 
    {help procrustes postestimation##procoverlay:{bf:[MV] procrustes postestimation}}.

{p 9 9 2}
    {cmd:predict} after {cmd:procrustes} produces fitted values for all
    variables, residuals for all variables, or residual sums of squares for a
    specified target variable; see 
    {help procrustes postestimation##predict:{bf:[MV] procrustes postestimation}}.

{p 5 9 2}
4.  New command {cmd:biplot} performs a biplot analysis of a dataset and
    produces a two-dimensional biplot of the results.  A biplot simultaneously
    displays the observations (rows) and the relative positions of the
    variables (columns).  Observations are projected to two dimensions such
    that the distance between the observations is approximately preserved.
    The variables are plotted as arrows, with the cosine of the angle between
    arrows approximating the correlation between the variables.
    See {helpb biplot:[MV] biplot}.

{p 5 9 2}
5.  New command {cmd:tetrachoric} computes a tetrachoric correlation
    matrix for a set of binary variables.  {cmd:tetrachoric} is
    documented in {bf:[R]} but will often be used in multivariate analyses;
    see {help tetrachoric:{bf:[R] tetrachoric}}.

{p 9 9 2} 
    {cmd:tetrachoric} results can be used in subsequent factor
    analyses or principal component analyses using the new
    {cmd:factormat} and {cmd:pcamat} commands.  See 
    {helpb factor:[MV] factor} and {helpb pca:[MV] pca}.

{p 5 9 2} 
6.  Existing command {cmd:canon} now allows analysis and presentation of
    more than one linear combination and has new options for reporting the raw
    or standardized coefficients and for reporting significance tests of the
    canonical correlations; see {help canon:{bf:[MV] canon}}.

{p 9 9 2}
    The following new {cmd:estat} commands work after {cmd:canon} and
    provide additional statistics and results:

{p 9 13 2}
    a.  {cmd:estat} {cmd:correlations} reports the correlations among all
        variables.

{p 9 13 2}
    b.  {cmd:estat} {cmd:loadings} reports the matrices of canonical loadings.

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

{p 5 9 2}
7.  Existing command {cmd:cluster dendrogram} has many new features,
    including horizontal dendrograms and the ability to label branch counts.
    The look of the graph can now be changed (titles, axes, colors, etc.);
    see {help cluster dendrogram:{bf:[MV] cluster dendrogram}}.

{p 5 9 2}
8.  The existing hierarchical cluster commands have new option
    {cmd:measure()} that specifies the proximity measure to use in
    computing dissimilarities between observations.  Any of 33
     measures
    may be specified; see 
    {help measure_option:{bf:[MV]} {it:measure_option}}.  Previously
    most of the measures were available under other option names; those
    options continue to work but are undocumented.  See 
    {help cluster:{bf:[MV] cluster}}.

{p 5 9 2}
9.  Existing command {cmd:cluster stop} has new option {cmd:varlist()}
    that specifies alternative variables to use when computing the
    stopping rules; see {help cluster stop:{bf:[MV] cluster stop}}.


{title:What's new:  Analysis of proximity matrices}

{pstd}
All of Stata's multivariate analysis facilities that rely on pairwise
comparisons of distance, similarity, dissimilarity, covariance, correlation,
or other proximity measures can now work directly with
proximity matrices that you compute or obtain from other sources.

{pstd}
Previously, all these facilities worked only with raw datasets.  The new
commands implement analyses on matrices.  They share the common ability to
accept either full matrices or vectors representing the lower or upper
triangle of a symmetric proximity matrix.

{p 4 9 2} 
10.  New command {cmd:clustermat} extends all of Stata's hierarchical
     clustering facilities to the analysis of matrices of a dissimilarity
     measure (sometimes called a distance or proximity measure).  This
     includes all seven linkage methods and the ability to create dendrograms
     of the results; see {help clustermat:{bf:[MV] clustermat}}.

{p 4 9 2} 
11.  New command {cmd:factormat} performs factor analysis on a matrix of
     correlations, extending all the new and previously available
     capabilities of the existing command {helpb factor:[MV] factor} 
     to precomputed
     matrices of correlations; see {help factor:{bf:[MV] factor}}.

{p 4 9 2} 
12.  New command {cmd:pcamat} performs principal component analysis on an
    existing correlation or covariance matrix; see 
    {help pca:{bf:[MV] pca}}.

{p 4 9 2}
13.  New {cmd:matrix} subcommand {cmd:dissimilarity} computes similarity,
    dissimilarity, or distance matrices using any of 19 
    proximity measures for continuous data 
    and 14 measures for binary data; see 
    {help measure_option:{bf:[MV]} {it:measure_option}}
     and see 
    {help matrix dissimilarity:{bf:[MV] matrix dissimilarity}}.


{title:What's new:  Factor and principal component analysis additions}

{p 5 5 2} 
In addition to allowing direct analysis of correlation and covariance matrices
using {helpb factormat} and {helpb pcamat}, Stata's factor analysis and
principal components analysis (PCA) methods have been expanded,
particularly through the addition of postestimation commands for reporting and
graphing results.

{p 4 9 2} 
14.  Command {cmd:factor} has new reporting option 
    {cmd:altdivisor}, that specifies
    the trace of the correlation matrix be used as the divisor for
    proportions, rather than the default (the sum of all eigenvalues).

{p 4 9 2}
15.  
    New {cmd:estat} commands for use after {cmd:factor} and
    {cmd:factormat} provide additional statistics and
    results:

{p 9 13 2}
    a.  {cmd:estat} {cmd:common} reports the correlation matrix of the common
        factors and is more of interest after oblique rotations.

{p 9 13 2}
    b.  {cmd:estat} {cmd:factors} reports model-selection criteria (AIC and
        BIC) over all the factors retained in an analysis.

{p 9 13 2}
    c.  {cmd:estat} {cmd:rotatecompare} reports the unrotated factor loadings
        next to the most-recent rotated loadings.

{p 9 13 2}
    d.  {cmd:estat} {cmd:structure} reports the factor structure -- the
        correlations between the variables and the common factors.

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

{p 4 9 2} 
16.  Existing command {cmd:pca} allows several new options:

{p 9 13 2}
        a.  Option {cmd:vce(normal)} computes the VCE of the eigenvalues and 
            eigenvectors, assuming multivariate normality.  

{p 13 13 2}
            This gives you access to many of Stata's postestimation facilities
            for analyzing estimation results, including tests of eigenvalue and
            eigenvector significance, tests of linear and nonlinear
            combinations ({helpb test:[R] test} and {helpb testnl:[R] testnl}),
            linear and
            nonlinear combinations with confidence intervals 
            ({helpb lincom:[R] lincom}
            and {helpb nlcom:[R] nlcom}), 
            and nonlinear predictions with confidence
            intervals ({helpb predictnl:[R] predictnl}).

{p 13 13 2}
	    {cmd:vce(normal)} also produces the ingredients for 
            adding confidence intervals to screeplots;