Following long tradition, we are informing Statalist first:
Stata 11 begins shipping Monday, July 27.
Orders are now being accepted at http://www.stata.com
Below are some highlights from the release.
----------------
Factor variables
----------------
Probably the highlight of the release is factor variables, if only because
everyone is going to be using them. Stata itself now deeply understands
factor variables. -xi- is dead. You can type
----------------------------------------------------------------------
. regress y i.sex i.group i.sex#i.region age (1)
. regress y i.sex##i.group age (same as 1)
. regress y i.sex i.group i.region
i.sex#i.group i.sex#i.region i.group#i.region (2)
i.sex#i.group#i.region
age
. regress y i.sex##i.group##i.region age (same as 2)
----------------------------------------------------------------------
Stata itself forms the necessary indicator variables. NO NEW VARIABLES
ARE CREATED IN YOUR DATA.
In the above, # means interaction, and ## means factorial interaction, so
A##B is equivalent to A B A#B. Asterisks would have looked better, but *
is Stata's varlist wildcard indicator, and factor variables are now just
part of varlists, so * and # had to coexist. Anyway, interactions are
much more like Kronecker products than like multiplication.
By the way, I typed -i.- everywhere above, but you can type, for example,
-sex#group- and Stata will know that you mean -i.sex#i.group-.
You can form interactions of factor variables with continuous variables,
and continuous variables with continuous variables, by using the -c.-
prefix:
----------------------------------------------------------------------
. regress y i.sex##i.group##i.region
age c.age#c.age (3)
. regress y i.sex##i.group##i.region
age i.sex##i.group##i.region#c.age (4)
c.age#c.age i.sex##i.group##i.region#c.age#c.age
. regress y i.sex##i.group##i.region##c.age (same as 4)
i.sex##i.group##i.region##c.age#c.age
. regress y i.sex##i.group##i.region##(c.age c.age#c.age) (same as 4)
----------------------------------------------------------------------
In the last example, note the clever use of # and ## together, with ##
being used to produce factorial-style interactions and # being used to
square age.
This new factor-variable notation is understood by all but a handful of
estimation commands. I demonstrated with -regress-, but I could have
used nearly any other estimation command.
----------------------------------------------------------------------
. logistic outcome i.treatment##i.sex age bp c.age#c.bp
----------------------------------------------------------------------
Factor variables work with -summarize- and -list-, too:
----------------------------------------------------------------------
. list outcome i.treatment##i.sex
----------------------------------------------------------------------
That is useful for understanding exactly what the notation produces.
--------
Graphics
--------
You can now put bold and italic text, Greek letters, symbols,
superscripts, and subscripts on graphs. What more is there to say?
----------
Statistics
----------
There are many new statistics in Stata 11, including
o multiple imputation
o competing-risks survival-time regression
o GMM estimation with user-specified moment functions
o new -margins- command, which replaces -mfx- and -adjust-, and does
so much more
o state-space modeling
o multivariate GARCH
o dynamic-factor models
o unit-root tests for panel data
o error structures for linear mixed models
o standard errors for BLUPs in linear mixed models
There is so much to say about multiple imputation that Stata's new -mi-
command gets its own manual.
See http://www.stata.com/stata11 for details.
------------------------------
Graphical user interface (GUI)
------------------------------
New GUI features include
o Variables Manager
Edit names, labels, display formats, storage types, notes, and
value labels. Those with many variables can use a filter to
see a selected subset of variables.
o Data Editor
Live view onto your data, filters, data snapshots, and more.
o Do-file Editor for Stata for Windows
Syntax highlighting, code folding, and no limit to file size.
I admit that I do not use GUI features often, but the new Variables
Manager does indeed make things easy. Others here tell me that the
new Data Editor will be the most popular interface feature because
you can now leave it open while you perform your analysis, and changes
are reflected instantly in it. You can even perform all your data
management from within the Data Editor, and do so in a reproducible
manner because it issues Stata commands for all changes.
-----------
Programming
-----------
Mata now includes full object-oriented programming facilities: classes,
inheritance, constructors and destructors, public/private/protected
declarations, virtual functions, and more. Just as in Java, code is
fully compiled, so there is no speed penalty for using it.
-----------
PDF manuals
-----------
Stata's manuals now ship in PDF format with every copy of Stata. All the
manuals. That's 8,500 pages.
The manuals are [GS], [U], [R], [D], [G], [P], [M], [I], [MI], [MV], [ST],
[SVY], [TS], and [XT].
Even better, the manuals are fully integrated with the help system, so you
can click from a help file and jump to the right manual, the right page!
I admit that it's convenient having the manuals on my laptop.
Of course, the manuals are available in printed form, too.
-----------------
One more thing...
-----------------
By the way, if you have a multicore computer, now might be a good time to
upgrade to Stata/MP. It's faster, and it's even faster in Stata 11.
There is much more, so visit http://www.stata.com/stata11
-- 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/
