Stata is a registered trademark of StataCorp LP , College Station, TX, USA, and the Stata logo is used with permission of StataCorp.
StataCorp社 国内正規販売代理店 インフォーマティック（株） [Back] [Home] [Bottom] お問い合わせ・注文
2011年6月27日、米国Stata社は、Stataの新しいバージョンStata 12を発表しました。
出荷は7月25日に開始されました。価格は、おおむね値下げされました。
すでに、2011年5月および6月にStata 11を購入されたお客様には、無償アップグレードが提供されます
ので（送料別）ご相談ください。
今回も広範囲にわたる重要な機能拡張が加えられています。
主なところをあげると、新しい機能として下記の機能が強化または追加されました。
We have just announced on Statalist that Stata 12 will begin
shipping
on July 25.
Stata 12 has many, many new features. Here are just a few:
o Structural equation modeling (SEM)
This is a complete implementation of SEM. There will be a
separate Structural Equation Modeling Reference Manual.
Included are path diagrams, standardized and unstandardized
estimates, modification indices, direct and indirect effects,
score tests, Wald tests, factor scores and other predictions,
goodness of fit, estimation with groups and tests of invariance,
seemingly unrelated regression (SUR), instrumental variables,
FIML and LIML estimation of simultaneous equations, maximum
likelihood estimation, GMM estimation, clustered data, survey
data, missing at random (MAR) data. All has been implemented for
big datasets and for big models.
We are excited about this addition. We feel that this feature
will open up another market for Stata. Although some
psychologists use Stata already, a lot did not because of the
absence of this capability. StataCorp is placing advertisements
in several journals for psychologists telling them about Stata
12
and SEM.
Although I mentioned psychologists in the above paragraph,
anyone
who fits linear regressions, multivariate regressions, seemingly
unrelated regressions, or simultaneous systems, or anyone who
uses generalized method of moments (GMM), should be interested.
SEM is a remarkably flexible framework.
o Contrasts and pairwise comparisons
Capabilities include linear and nonlinear models, comparison
of
means, intercepts, or slopes, comparison of adjacent categories,
comparison with reference category, comparison with grand mean,
treatment effects, orthogonal polynomials, adjustments for
multiple comparisons, and graphs.
The contrast and pwcompare commands are available after
estimation commands that allow factor variables, which means
almost every estimation command in Stata.
o Multiple imputation
Many new additions have been made to multiple imputation. The
MI
manual grew from 270 pages to 353.
New features include chained equations, imputation of continuous,
ordinal, cardinal, and count variables, conditional imputation,
imputation separately within groups, support for panel data and
multilevel models, and linear and nonlinear predictions.
o Time series
This is another area where there is a lot that is new. The
time-series manual grew from 537 to 710 pages. New features
include the following:
Multivariate GARCH, including constant conditional correlations,
dynamic conditional correlations, varying conditional
correlations, level predictions, variance predictions, and
dynamic forecasts.
ARFIMA, featuring long memory processes, predictions, fractional
integration predictions, and dynamic forecasts.
UCM -- unobserved-components models -- including trend, seasonal,
and cyclical components, prediction of components, and dynamic
forecasts.
Spectral density with parametric estimates after ARIMA, ARFIMA,
and UCM.
Time-series filters, including Christiano-Fitzgerald, Baxter-King
band-pass filters, Hodrick-Prescott high-pass filters, and
Butterworth high-pass filters.
Business calendars. This is really great. Users can define
their own calendars so that dates display correctly and lags and
leads work as they should. For instance, a user might create a
file that recorded the days the London Stock Exchange is open.
Say that file was named lse.stbcal, Stata would then understand
the format %tblse, just as it understands the usual date format
%td.
o Contour plots, including both filled and outlined plots
o Multilevel/mixed models. Estimation with complex survey data
is
now possible with xtmixed, including multilevel sampling weights
and robust variance estimators. This puts Stata at the forefront
of handling survey data in linear mixed models.
o ROC analysis capabilities have been greatly expanded. ROC
is
documented in the Base Reference Manual. Documentation went from
32 pages to 121. New features include ROC adjusted for
covariates, which is to say, you can model sensitivity and
specificity of a diagnostic test to take into account prognostic
factors such as age and gender. You can also easily graph the
estimated ROC curves that result from these models.
o New commands import excel and export excel read and write
worksheets from Microsoft Excel files, .xls and .xlsx. And,
there is a new Import Preview tool which lets you see the file's
contents and adjust import settings before you import the data.
We have also added support for importing EBCDIC files.
o PDF export of results and graphs
o Interface enhancements
Stata's interface has many new features and enhancements. The
screen layout has been redesigned to better fit wide screens.
There is a new Properties window which lets you manage your
variables, including their names, labels, value labels, notes,
formats, and storage types.
The Data Editor also has a new Properties window; has another
tool that lets you Hide, Show, Filter, and Reorder the variables;
and the new Clipboard Preview tool lets you see and prepare your
raw data before pasting.
The Review and Variables windows support filtering (searching),
and you can now also search in the Results window.
o Automatic memory management
This is one of those features that is difficult to appreciate
before you actually use it, but I predict that users are going
to
love this. No more setting of memory. Stata automatically
adjusts its memory usage up and down according to current
requirements.
o Improved help files
This is another feature that is difficult to appreciate before
using it, but it is really nice. The Viewer shows help files with
pulldown menus for the dialog box, the Also See entries, and a
"Jump to" pulldown that includes each of the main sections
in the
help file so users can go quickly to the information that they
want to see. For instance, a user could click on "Examples"
in
the Jump to pulldown to go directly to the examples for the
command.
There are many more new features including more improvements
in the
Stata interface, Data-Editor, and Do-file Editor, plus more parallelized
commands and speed ups in Stata/MP versions.
As you probably can tell, I am excited about Stata 12. I have
seen and
used the new features and believe that our users are going to
greet the
announcement with lots of enthusiasm.
There is a slight change to the Stata 12 dataset format due
to support
for business calendar formats. However, in the final free update
to
Stata 11, we are including support for reading Stata 12 datasets
in
Stata 11. And, as always, Stata 12 can read datasets from all
prior
versions of Stata.
The new version of Stat/Transfer (version 11, coming soon)
supports the
Stata 12 dataset format. The previous version of Stat/Transfer
(version
10) will work for transferring data to Stata 12, but not for
transferring data from Stata 12 to another format. I will let
you know
when I have information on the exact availability date for Stat/Transfer
version 11.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
StataCorp 1-800-782-8272 U.S.
4905 Lakeway Drive 1-800-248-8272 Canada
College Station TX 77845 1-979-696-4600 Intl
1-979-696-4601 Fax
VISIT OUR WEB SITE http://www.stata.com
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
[Top]
[Home] [Top] [Back] お問い合わせ・注文