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

{smcl}
{* 03may2005}{...}
{cmd:help whatsnew8to9}
{hline}

{title:Title}

{pstd}
{bf:[U] 1.3 What's new}


{title:Description}

{pstd}
What is changed between Stata 8 and Stata 9 is described.

{pstd}
This information is intended for previous Stata users.  If you are new to
Stata, you may as well skip it.

{pstd}
See {help whatsnew} for the updates made since the release of Stata 9.


{title:Remarks}

{pstd}
Some of the important new additions include

{p 8 12 2}
1.  New matrix programming language Mata.

{p 8 12 2}
2.  New survey features, including 
    balanced repeated replications (BRR)
    and jackknife variance estimates, 
    complete support for multistage designs, 
    and poststratification.

{p 8 12 2}
3.  Estimation of linear mixed models, including standard errors and 
    confidence intervals for all variance components.

{p 8 12 2}
4.  Estimation of multinomial probit models, 
    including support for several correlation structures and for 
    user-defined structures.

{p 8 12 2}
5.  New multivariate analysis, including multidimensional scaling, 
    correspondence analysis, and Procrustean analysis, along with 
    the ability to analyze proximity matrices as well as raw data.

{p 8 12 2}
6.  Improved GUI, including multiple Do-file Editors, multiple Viewers,
     and multiple Graph windows; multiple windowing preferences; dockable
     windows; and much more.

{pstd}
There are other major features, and it will take us another 30 pages to
mention everything.

{pstd}
What's new is presented under the headings 

		{helpb whatsnew8to9##mata:New matrix language}

		{helpb whatsnew8to9##survey:Survey statistics}

		{helpb whatsnew8to9##panel:Longitudinal/panel data}
		
		{helpb whatsnew8to9##timeseries:Time-series statistics}

		{helpb whatsnew8to9##multivariate:Multivariate statistics}

		{helpb whatsnew8to9##survival:Survival analysis}

		{helpb whatsnew8to9##general:General-purpose statistics}

		{helpb whatsnew8to9##ml:New ML features}

		{helpb whatsnew8to9##functions:Functions and expressions}

		{helpb whatsnew8to9##data:Data management}

		{helpb whatsnew8to9##graphics:Graphics}

		{helpb whatsnew8to9##gui:User interface}

		{helpb whatsnew8to9##programming:Programming}

		{helpb whatsnew8to9##documentation:Documentation}


{marker mata}{...}
{title:What's new:  New matrix language}

{pstd}
Stata has an all-new matrix language called Mata, which is the subject
of its own manual, {bf:[M] The Mata Reference Manual}.
Mata can be used by those who want to think in matrix terms and perform 
matrix calculations interactively, and it can be used by
programmers who want to add features to Stata.

{pstd}
Mata has been used to implement many of the new features found in this
release.  Mata is compiled, optimized, and fast.

{pstd}
Stata's previously existing {cmd:matrix} command continues 
to be documented.  There is an admittedly uneasy relationship between
the two, but {cmd:matrix} continues to have its uses.  For
serious computation, however, you will definitely want to use the new
language.

{pstd}
See {helpb mata:[M-0] intro} -- or {cmd:help} {cmd:mata} -- which provides an
introduction and organized reading list.  The first thing you will read is
{helpb m1_first:[M-1] first}.

 
{marker survey}{...}
{title:What's new:  Survey statistics}

{pstd}
Stata 9 substantially extends Stata's survey-analysis and
correlated-data-analysis facilities by adding the remaining two methods of
computing standard errors -- Balanced Repeated Replications (BRR) and survey
jackknife.

{pstd}
Stata 9
also adds complete support for multistage sampling and poststratification.  

{pstd}
A new, unified syntax is used for declaring the design of survey data and for
fitting models.  For an overview of all survey facilities, see
{help survey:{bf:[SVY] survey}}.

{pstd}
All the old syntax continues to work under version control, although the
survey estimation commands do not require that, but if you use old
syntax, the new features will not be available.

{p 5 9 2} 
1.  Existing command {cmd:svyset} for declaring the survey
    design has new syntax that supports a host of new features in Stata's
    survey-analysis facilities:  

{p 9 13 2}
    a.  BRR and jackknife variance estimators have been added to the
        previously available linearization variance estimator.
        Moreover, use of BRR or jackknife (or linearization) can now be 
        specified when you {cmd:svyset} or at estimation time.
    
{p 9 13 2}
    b.  Multistage designs can now be declared, and they may have primary, 
        secondary, and lower-stage sampling units.  The linearization variance
	estimator takes complete advantage of the information in multistage
	designs.

{p 9 13 2}
    c.  Stratification is now allowed in all stages, making variance estimates
        more efficient wherever stratification can be exploited.

{p 9 13 2}
    d.  Poststratification is now available and, like stratification,
        also makes variance estimates more efficient.  Poststratification
        adjusts weights, improves variance estimates, and accounts for biases
        when demographic or other groupings are known.

{p 9 13 2}
    e.  Finite-population corrections are now allowed in all stages.

{p 9 13 2}
    f.  Sampling weights are handled under all three variance estimators.

{p 9 9 2} 
    For details, see {manhelp svyset SVY}.
    The previous {cmd:svyset} syntax continues to work under version control.

{p 5 9 2}
2.  New prefix command {cmd:svy:} is
    how you tell estimators you have survey data.  You no longer type
    {cmd:svyregress}; you type {cmd:svy: regress}.  This is not just a matter
    of style; {cmd:svy} really is a prefix command, and in fact, you can even
    use it as a prefix on estimation commands you write.  In addition,
    {cmd:svy:} provides a standard, unified syntax for accessing Stata's 
    survey features.  {cmd:svy:} is easy to use because it automatically
    applies everything you have previously {cmd:svyset}, including the design.

{p 9 9 2} 
    The following estimators can be used with {cmd:svy:} prefix:

{p2colset 14 36 38 2}{...}
{p 9 9 2}
{bf:Descriptive statistics}

{p2col :{helpb "svy: mean"}}Population and subpopulation means{p_end}
{p2col :{helpb "svy: proportion"}}Population and subpopulation proportions{p_end}
{p2col :{helpb "svy: ratio"}}Population and subpopulation ratios{p_end}
{p2col :{helpb "svy: total"}}Population and subpopulation totals{p_end}

{p2col :{helpb "svy: tabulate oneway"}}One-way tables for survey data{p_end}
{p2col :{helpb "svy: tabulate twoway"}}Two-way tables for survey data{p_end}

{p 9 9 2}
{bf:Regression models}

{p2col :{helpb "svy: regress"}}Linear regression{p_end}
{p2col :{helpb "svy: ivreg"}}Instrumental variables regression{p_end}
{p2col :{helpb "svy: intreg"}}Interval and censored regression{p_end}

{p2col :{helpb "svy: logistic"}}Logistic regression, reporting odds ratios{p_end}
{p2col :{helpb "svy: logit"}}Logistic regression, reporting coefficients{p_end}
{p2col :{helpb "svy: probit"}}Probit regression{p_end}

{p2col :{helpb "svy: mlogit"}}Multinomial logistic regression{p_end}
{p2col :{helpb "svy: ologit"}}Ordered logistic regression{p_end}
{p2col :{helpb "svy: oprobit"}}Ordered probit models{p_end}

{p2col :{helpb "svy: poisson"}}Poisson regression{p_end}
{p2col :{helpb "svy: nbreg"}}Negative binomial regression{p_end}
{p2col :{helpb "svy: gnbreg"}}Generalized negative binomial regression{p_end}

{p2col :{helpb "svy: heckman"}}Heckman selection model{p_end}
{p2col :{helpb "svy: heckprob"}}Probit model with selection{p_end}

{p 9 9 2} 
    Previously existing survey-estimation commands, such as {cmd:svyregress},
    {cmd:svymean}, and {cmd:svypoisson}, continue to work as they did before,
    but only if your survey design is declared using {cmd:version 8: svyset}
    or if you are working with an old Stata 8 dataset.  For a mapping from old
    estimation commands to the new syntax, see {help svy8}.  (The new
    prefix {cmd:svy:} works with datasets that were {cmd:svyset} under an
    earlier release of Stata.)

{p 9 9 2} 
    In addition to the three variance estimators and support for multistage
    sampling, the new {cmd:svy:} prefix provides other enhancements, including

{p 9 13 2}
    a.  Option {cmd:subpop()} allows more
        flexible selection of subpopulations, meaning that more general
        {cmd:if} conditions are now allowed.

{p 9 13 2}
    b.  Strata with only one sampling unit (sometimes called singleton PSUs)
        are now handled better -- the coefficients are now reported, but with
        missing standard errors.  {cmd:svydes} can now be used to find and
        describe these strata;
        see {manhelp svydes SVY}.

{p 9 13 2}
    c.  With BRR variance estimation, a Hadamard matrix can be used in
        place of BRR weights, and Fay's adjustment may be specified;
        see {manhelpi brr_options SVY}.

{p 5 9 2}
3.  New command {cmd:svy:} {cmd:proportion} replaces {cmd:svyprop}.
    (By the way, new command {cmd:proportion} can be used without the 
    {cmd:svy:} prefix; see {manhelp proportion R}.)
    Unlike {cmd:svyprop}, {cmd:svy:} {cmd:proportion} is an estimation
    command and computes a full covariance matrix for all of the estimated
    proportions, allowing postestimation features, such as tests of linear and
    nonlinear combinations of proportions ({helpb test} and {helpb testnl}) or
    creation of linear and nonlinear combinations with confidence intervals
    ({helpb lincom} and {helpb nlcom}).

{p 5 9 2}
4.  New commands {cmd:ratio}, {cmd:total}, and
    {cmd:mean}, used with the {cmd:svy:} prefix, 
    use casewise deletion and estimate full
    covariance matrices for the estimates.  

{p 5 9 2}
5.  New command {cmd:svy: tabulate oneway} addresses a missing feature.
    Previously, anyone wanting a one-way tabulation had to create a constant
    and perform two-way survey tabulation with that constant.

{p 5 9 2}
6.  New command {cmd:estat} computes and reports additional statistics and
    information after estimation with {cmd:svy:} prefix:

{p 9 13 2}
    a.  {cmd:estat} {cmd:svyset} reports complete information on the survey
        design.

{p 9 13 2}
    b.  {cmd:estat} {cmd:effects} computes and reports the design
        effects -- DEFF and DEFT -- and the misspecification effects -- MEFF
        and MEFT -- in any combination for each estimated parameter.

{p 9 13 2}
    c.  {cmd:estat} {cmd:effects} can also compute DEFF and DEFT for
        subpopulations using simple random-sample estimates from either the
        overall population or from the subpopulation.  {cmd:estat}
	{cmd:effects} replaces and extends the {cmd:deff}, {cmd:deft},
	{cmd:meff}, and {cmd:meft} options previously available on survey
	estimators.

{p 9 13 2}
    d.  {cmd:estat} {cmd:lceffects} computes and reports the survey design
	effects and misspecification effects for any linear combination of
	estimated parameters.

{p 9 13 2}
    e.  {cmd:estat} {cmd:size} reports the sample and population sizes for
        each subpopulation after {cmd:svy:} {cmd:mean}, 
	{cmd:svy:} {cmd:proportion}, {cmd:svy:} {cmd:ratio}, 
	and {cmd:svy:} {cmd:total}.

{p 9 9 2}
     For details on {cmd:estat} after survey estimation,
     see {help svy estat:{bf:[SVY] estat}}.

{p 5 9 2}
7.  Existing command {cmd:svydes} has several new features and options:

{p 9 13 2}
	a.  New option {cmd:stage()} lets you select the sampling stage
	    for which sample statistics are to be reported.

{p 9 13 2}
	b.  New option {cmd:generate()} identifies strata with a single
	    sampling unit.

{p 9 13 2}
	c.  New option {cmd:finalstage} replaces {cmd:bypsu} and reports
            observation sample statistics by sampling unit in the final stage.