Re: st: slide presentation by stata employees

["Michael S. Hanson" <mshanson@mac.com>]

On Dec 16, 2006, at 2:04 PM, Christer Thrane wrote:

> Sorry for being very ignorant here, perhaps it's because I am Word, 
> Excel and PowerPoint guy (In additiaon to Stata, of course).
>
> What do I have to buy/dowloade/whatever of software to produce slides 
> like these, and where can I find manuals to help me?
	(Deep breath...)  OK.  Here are the basics as I see them.  Others will 
likely chime in helpfully with their own takes on this question.

	In a nutshell, you will need to install and learn to use LaTeX in 
order to make slides like those referenced by the OP.  This is not 
something you can do in a day, or even a week.  Briefly, here's how to 
tell if making the plunge into LaTeX is right for you:


	* If you use Stata primarily with the GUI and/or a small set of 
commands you have committed to memory (or have a "cheat sheet" of 
commands to type), or tend to use SPSS or EViews as much as Stata, then 
you should stick with the "Office tools" you currently use.


	* If you use Stata exclusively from the command line, and are 
comfortable writing .do files of moderate complexity (e.g., loops, 
macros), then you might be a candidate to use LaTeX.  (So too if you 
can move more-or-less smoothly between programming in Stata and another 
statistical package like SAS, TSP or RATS.)  Be aware LaTeX is 
non-trivial to install, maintain, and learn -- more so than Stata on 
each of these dimensions.  There are numerous distributions available 
for all major computing platforms that have more-or-less automated 
installation -- but in the (relatively rare) event something goes 
wrong, it can be incredibly difficult for a non-TeXpert to diagnose.  
There also are numerous books and on-line reference materials, should 
you chose to invest the time into learning how to code in LaTeX.  If 
you can grok using programming commands to write papers and make 
slides, much like you use programming commands to undertake statistical 
analysis in Stata, then LaTeX might be for you.  Take a look at, for 
example:

"Getting to Grips with LaTeX":
	<http://www.andy-roberts.net/misc/latex/>
"The Not-So-Short Introduction to LaTeX":
	<http://tobi.oetiker.ch/lshort/lshort.pdf>


	* If you regularly program your own Stata commands in sophisticated 
.ado files, or find yourself torn between using Stata and R, then get 
thee to <http://www.ctan.org/> immediately!  Why are you wasting time 
with anything else?!


	Note that there exist (La)TeX editors and tools that attempt to place 
a GUI-like interface on the coding intrinsic with LaTeX;  LyX is one 
such example.  Having never used those tools extensively (I use TeXShop 
or BBEdit to code in LaTeX), I do not know whether they can be used to 
produce presentations with beamer.

	LaTeX is very powerful, cross-platform, and very attractive to those 
who care about how their work looks in print -- specifically, 
mathematical typography.  (As free software, it also attracts people 
who are morally opposed to MS -- you appear to not be one of them.)  If 
you are simply interested in getting stuff in print quickly with 
minimal fuss, and are content if the output is "good enough," then 
frankly I doubt you will see the advantage to using LaTeX, and likely 
will find it extremely difficult to motivate yourself to learn it.

	Two other considerations:  First, if you have someone at your current 
institution with LaTeX experience whom you can consult, you will find 
learning LaTeX will be significantly easier than if you attempt to do 
it on your own.  Second, if file compatibility with colleagues and 
co-authors who are, for whatever reason, MS Office devotees is 
important to you, then you almost certainly should not bother with 
LaTeX and beamer.

	I hope that synopsis was helpful;  I attempted to be as honest and 
neutral in my assessment as possible.  As noted above, others may have 
different opinions.  Like many on Statalist, I find using LaTeX to be a 
desirable means to express my research in writing.  But it is not the 
best (nor even a good) tool for everyone.  As they say, YMMV.

                                        -- Mike

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/