Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Stata 12 Announcement


From   David Souther <[email protected]>
To   [email protected]
Subject   Re: st: Stata 12 Announcement
Date   Sun, 26 Jun 2011 21:20:36 -0500

Very exciting, but really hope theyve upped the limit in the number of
letters allowed in a string variable (and variable label) - especially
since they are allowing for more import/export to excel (which
supports much, much longer strings in cells).

DS


On Mon, Jun 27, 2011 at 2:54 AM, John Antonakis <[email protected]> wrote:
> Fantastic; particularly the SEM module. It's really helpful to be able to do
> simultaneous equations with latent variables in Stata. I hope the Stata SEM
> module will be developed further (the benchmark SEM program at this point
> being MPlus, though you probably are a step ahead in some aspects, or could
> be, e.g., limited information estimation of structural equation models using
> 2SLS, Hausman tests, etc.; if those capabilities are not in there now I
> really hope that they will be added too someday)!
>
> Well done, Stata! Really well done!
>
> John.
>
> __________________________________________
>
> Prof. John Antonakis
> Faculty of Business and Economics
> Department of Organizational Behavior
> University of Lausanne
> Internef #618
> CH-1015 Lausanne-Dorigny
> Switzerland
> Tel ++41 (0)21 692-3438
> Fax ++41 (0)21 692-3305
> http://www.hec.unil.ch/people/jantonakis
>
> Associate Editor
> The Leadership Quarterly
> __________________________________________
>
>
> On 27.06.2011 01:07, William Gould, StataCorp LP wrote:
>>
>> Following long tradition, we are informing Statalist first:
>>
>>     Stata 12 begins shipping Monday, July 25.
>>
>>     Orders are now being accepted at http://www.stata.com.
>>
>> Below are some highlights.
>>
>>
>> ---------------------------
>> Automatic memory management
>> ---------------------------
>>
>>     Automatic memory management means that you no longer have to
>>     -set memory- and never again will you be told that there is no
>>     room because you set too little!  Stata automatically adjusts its
>>     memory usage up or down according to current requirements.
>>
>>     The memory manager is tunable.  You can set a maximum if you wish.
>>     Old do-files can still -set memory-.  Stata merely responds, "-set
>>     memory- ignored".
>>
>>     We have tested the memory manager on systems with 1 TB (the largest
>>     currently available), and it is designed to scale to even more
>>     memory.
>>
>>
>> -----------------------------------------------------------
>> Import Excel files, export PDFs, and new interface features
>> -----------------------------------------------------------
>>
>>     Importing Excel files is easy.  And the new Import Preview Tool
>>     lets you see the file's contents and adjust import settings before
>>     you import it.
>>
>>     You can now directly export PDFs of graphs and logs.
>>
>>     Stata's windows are now laid out to fit wider screens better.  You
>>     can still get back the old layout from Edit ->  Preferences.
>>
>>     A new Properties window -- always available -- lets you manage
>>     your variables, including their names, labels, value labels,
>>     notes, formats, and storage types.
>>
>>     The Viewer is now tabbed, and it has buttons at the top to access
>>     dialogs, to jump within the document, and to jump to Also See
>>     documents.
>>
>>     The Data Editor also has a new Properties window; has another tool
>>     that lets you Hide, Show, Filter, and Reorder the variables; and
>>     has the new Clipboard Preview tool, which lets you see and prepare
>>     your raw data before pasting.
>>
>>
>> ----------------------------------
>> Structural equation modeling (SEM)
>> ----------------------------------
>>
>>     -sem- is a new estimation command, itself the subject of
>>      an entire manual.
>>
>>     If you are new to SEM, you should be interested if you fit linear
>>     regressions, multivariate regressions, seemingly unrelated
>>     regressions, or simultaneous systems, or if you're interested in
>>     generalized method of moments (GMM).  And if you think you are
>>     still not interested, take a look anyway.  SEM is a remarkably
>>     flexible framework.
>>
>>     If you know about SEM, you will be more interested in path
>>     analysis models, single- and multiple-factor measurement models,
>>     MIMIC models, latent growth models, correlated uniqueness models,
>>     and more, all of which can be fit by -sem-.  You will also be
>>     interested in -sem-'s standardized and unstandardized coefficients,
>>     direct and indirect effects, goodness-of-fit statistics,
>>     modification indices, predicted values and factor scores, and
>>     groupwise analysis with tests of invariance.
>>
>>     You can use the GUI or command language to specify your model.
>>     The command language is a variation on standard path notation.
>>     You can type
>>
>>         . sem (L1 ->  m1 m2 m3)
>>               (L2 ->  m4 m5)
>>               (L1 ->  L2)
>>
>>     In -sem-, lowercase names refer to variables in the data and
>>     uppercase names are latent variables.  The above corresponds to
>>
>>               m1 = a1 + b1*L1 + e1
>>               m2 = a2 + b2*L1 + e2
>>               m3 = a3 + b3*L1 + e3
>>
>>               m4 = a4 + b4*L2 + e4
>>               m5 = a5 + b5*L2 + e5
>>
>>               L2 = c1 + d1*L1 + e6
>>
>>     Maximum likelihood (ML) and asymptotic distribution free (ADF)
>>     estimation methods are provided.  ADF is generalized method of
>>     moments (GMM).  Robust estimates of standard errors and SEs for
>>     clustered samples are available, as is full support for survey
>>     data via the -svy:- prefix.  Missing at random (MAR) data are
>>     supported via FIML.
>>
>>
>> ----------------------------------------
>> Survey, cluster robust, and mixed models
>> ----------------------------------------
>>
>>     -xtmixed- now supports sampling weights and robust and cluster-
>>     robust standard errors for use with survey data, although you do
>>     *NOT* use the -svy:- prefix as you might have expected.
>>
>>     That is because multilevel models with survey data differ from
>>     standard models in that sampling weights need to be specified at
>>     each modeling level rather than just at the observation level.
>>     Sampling weights must reflect selection probability conditional on
>>     selection at the next highest level.
>>
>>     Thus, -xtmixed- expects you to specify a weight for each level in
>>     your model and warns you if you do not.
>>
>>
>> -------------------
>> Multiple imputation
>> -------------------
>>
>>     -mi impute- now supports
>>
>>         1.  Chained equations.
>>             Chained equations are used to impute missing values when
>>             variables may be of different types and missing-value
>>             patterns are arbitrary.  The first variable could be
>>             imputed using logit, the second using linear regression,
>>             and the third using multinomial logistic regression.
>>
>>         2.  Conditional imputation.
>>             Conditional imputation is customized imputation within
>>             group when group itself might be imputed.  You can
>>             restrict imputation of number of pregnancies to females
>>             even when female itself contains missing values and so is
>>             being imputed.
>>
>>         3.  Imputation by groups.
>>             Australians could have their missing values imputed using
>>             data from other Australians only.
>>
>>     -mi estimate- now
>>
>>         1.  Supports panel-data and multilevel models, so you can use
>>             -mi- with -xtreg- or -xtmixed-.
>>
>>         2.  Allows you to measure the amount of simulation error in
>>             your final model, so you can decide whether you need more
>>             imputations.
>>
>>     -mi predict- and -mi predictnl- create linear and nonlinear
>>     predictions in the original (m=0) data, and not just for complete
>>     observations but also for observations with missing values.
>>
>>
>> -----------
>> Time series
>> -----------
>>
>>     Check out the
>>
>>         1.  New estimators for
>>                 a.  GARCH
>>                 b.  ARFIMA
>>                 c.  UCM
>>
>>         2.  New postestimation command -psdensity- to estimate the
>>             spectral density of a stationary process using the
>>             parameters of a previously estimated parametric model.
>>
>>         3.  New command -tsfilter-, which filters a series to keep only
>>             selected periodicities (frequencies) and which can be used
>>             to separate a series into trend and cyclical components.
>>
>>     Multivariate GARCH deals with models of time-varying volatility in
>>     multiple series.  These models allow the conditional covariance
>>     matrix of the dependent variables to follow a flexible dynamic
>>     structure and the conditional mean to follow a
>>     vector-autoregressive (VAR) structure.
>>
>>     ARFIMA is a generalization of the ARMA and ARIMA models.  ARMA
>>     models assume short memory.  ARIMA models assume shocks are
>>     permanent.  ARFIMA provides the middle ground.  ARFIMA stands for
>>     autoregressive, fractionally integrated moving average.
>>
>>     UCM stands for unobserved component model and decomposes a series
>>     into trend, seasonal, cyclic, and idiosyncratic components after
>>     controlling for optional exogenous variables.
>>
>>
>> ------------------
>> Business calendars
>> ------------------
>>
>>     There is a new %t format:  %tb.  The b stands for business
>>     calendars.  Business calendars allow you to define your own
>>     calendars so that dates display correctly and lags and leads work
>>     as they should.
>>
>>     You could create file lse.stbcal that records the days the London
>>     Stock Exchange is open (or closed) and then Stata would understand
>>     format %tblse just as it understands the usual date format %td.
>>
>>     Once you define a calendar, Stata deeply understands it.  You can,
>>     for instance, easily convert between %tblse and %td values.
>>
>>
>> -----------------------------------
>> Constrasts and pairwise comparisons
>> -----------------------------------
>>
>>     We were tempted to call this "Stata for Experimentalists" except
>>     that the features are useful to Stata users of all disciplines.
>>
>>     Contrasts, pairwise comparisons, and margins plots are about
>>     understanding and communicating results from your model.  How does
>>     a covariate affect the response?  Is the effect nonlinear?  Does
>>     the effect depend on other covariates?
>>
>>     New commands -contrast-, -pwcompare-, and -marginsplot- join
>>     -margins-.
>>
>>         1.  -contrast- compares effects of factor variables and their
>>             interactions.  It can perform ANOVA-style tests of main
>>             effects, simple effects, interactions, and nested effects.
>>             It also decomposes these effects into comparisons against
>>             reference categories, comparisons of adjacent levels,
>>             comparisons against the grand mean, orthogonal
>>             polynomials, and such.
>>
>>             In addition to predefined standard contrasts, user-defined
>>             contrasts are also supported.  Consider
>>
>>                  . contrast ar.educ
>>
>>             The -ar.- out front is one of the new, predefined contrast
>>             operators.  -ar.- stands for "adjacent, reversed", and
>>             -contrast ar.educ- compares adjacent levels of education,
>>             for instance, high school to some college, some college to
>>             college graduate, etc.
>>
>>         2.  -pwcompare- performs all (or subsets) of the pairwise
>>             comparisons.  This can be done for all levels of a single
>>             factor variable or for interactions or interactions with
>>             continuous variables.
>>
>>         3.  -margins- now allows the new contrast operators and has a
>>             -pwcompare- option to perform pairwise comparisons.
>>
>>         4.  -marginsplot- graphs results from -margins-.
>>
>>
>> ---------------------------
>> ROC adjusted for covariates
>> ---------------------------
>>
>>     New command -rocreg- is like regression for ROC.  You can model
>>     how sensitivity and specificity depend on covariates, and you
>>     can draw graphs.
>>
>>
>> -------------
>> Contour plots
>> -------------
>>
>>      You just have to see one.  Visit
>>      http://www.stata.com/stata12/contour-plots/
>>
>>
>> ----
>> More
>> ----
>>
>>     There's more.  For instance -rename- has a new syntax that allows
>>     you to rename groups of variables.
>>
>>         . rename (vara varb varc) (varc varb vara)
>>
>>     swaps the names around.
>>
>>         . rename jan* *1
>>
>>     renames all variables starting with jan to instead end in 1.
>>
>>         . rename v# stat#
>>
>>     renames v1 to be stat1, v2 to be stat2, and so on.
>>
>>         . rename v# v(##)
>>
>>     renames v1 to be v01, v2 to be v02, ...
>>
>>         . rename (a b c) v#, addnumber
>>
>>     rename a to be v1, b to be v2, and c to be v3.
>>
>>         . rename v# (a b c)
>>
>>     does the reverse.
>>
>>
>>
>> There really is a lot more.  See http://www.stata.com/stata12.
>>
>>
>> -- Bill
>> [email protected]
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index