[R] Statistical Software Comparison

[Robert Duval]

I'll answer the ones I know of:

> 4. Manuals (theory included in the manuals).
Stata manuals are superb. The online help manuals are really minimal,
but the complete set of manuals for sale is really good. Not only they
discuss the Stata implementetation, but they give a concise
theoretical discussion of what the statistical methods are actually
doing.

While they don't get to talk much about the inner workings of Stata,
(as some of the R manuals do) I like them much better to the R ones.

Many of the statistical procedures are illustrated with examples using
the datasets included with the software

> 5. Support (in this aspect there is no comparison with R,
>     the R list is the best known support).

R list has a better support than Statalist, but still Statalist is
quite active and helpful.
Plus they are more polite... no RTFM or stuff like that.
If you own a Stata license, you can get direct support from somebody
at StataCorp (in addition to Statalist). This is specially relevant if
you have questions on how Stata is estimating something, bugs, etc.


> 6. Numerical stability.
Quite stable. The only glitch I've observed is that after new releases
their routines are not very reliable... meaning they sometimes change
the way something is being computed and might they mess up something
that previously was running fine.

Right now Stata 9 is pretty stable, but if Stata 10 would come up in
the market now, I would probably wait for a couple of months and make
sure everything is well tested.


One last thing, while I abandoned the Stata world to move to R (due to
$$), I have to say that the only thing I really miss about it is its
ability to handle large datasets. Stata comes with great Data
management routines, and it can hold large amount of data in its
memory.
Here R is light years behind. This is particularly relevant if you
have to "clean-up" large datasets before you actually start doing
statistics.

hope this helps
robert

On 11/21/06, Kenneth Cabrera <krcabrer at une.net.co> wrote:
> Hi R users:
>
> I want to know if any of you had used
> Stata or Statgraphics.
>
> What are the advantages and disadvantages with
> respect to R on the following aspects?
>
> 1. Statistical functions or options for advanced
>     experimental design (fractional, mixed models,
>     greco-latin squares, split-plot, etc).
> 2. Bayesian approach to experimental design.
> 3. Experimental design planing options.
> 4. Manuals (theory included in the manuals).
> 5. Support (in this aspect there is no comparison with R,
>     the R list is the best known support).
> 6. Numerical stability.
> 7. Implementation of modern statistical approaches.
>
> Thank you for your help.
>
> Kenneth
>
> --
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>