编程较多的是R吧,stata要求没那么高

https://bbs.pinggu.org/thread-349743-1-1.html
看10、ucla提供的对sas，stata，spss，r的技术报告，比较全面
看完就知道了
软件的比较，一两句是很难说清楚的。那个技术报告全面介绍了这几个软件的差异和优劣。

下面贴出一段



Stata and SAS
    Stata and SAS complement each other nicely in terms of their features. Where one of the packages might
show a weakness, the other package complements it with a strength.

SAS is weak in terms of survey data analysis, but Stata is very strong in this area.
Stata is not good with extremely large (1+ or 2+ gigabyte) data files, but SAS is strong with such files.   
It is diffcult to use SAS for bootstrap, jacknife or Monte Carlo methods, but Stata excels in these areas.
Stata is not very good at reading hierarchical or complex raw data files, but SAS is very good at this.
SAS is not as good at handling weights, while Stata is very good at handling weights.

The major omissions I would see would be in the area of reading foreign file formats
(which could be remedied by purchasing Stat/Transfer) and that SPSS is stronger at some kinds of tasks
in ANOVA than SAS. But, overall, when you consider the coverage that you get with these two packages
combined, it is extremely impressive the breadth and depth of data management and statistical features that
these two packages, in concert, provide.

SAS and SPSS
SAS and SPSS do not complement each other very well at all. Both packages are weak in the analysis of
survey data and neither o®ers the interpretive tools to help interpret logit models in terms of predicted
probabilities. Neither package is strong with bootstrap/jacknife and monte carlo methods { they can do
these tasks, but can be di±cult to use. There are many special purpose regression models that may be either
unsupported or di±cult to implement. Neither is as strong as Stata with respect to weights. There are
models that cannot be estimated with clustering and/or such models would be di±cult to specify. Neither
package has tools for exporting data to special purpose statistical packages. Combining these two tools in
your toolkit leaves a number of weaknesses.

Stata and SPSS
Stata and SPSS do not complement each other very well. While SPSS adds its strengths of reading foreign
¯le formats and ANOVA to what Stata can do, it does not o®er much more in terms of the what I have seen
clients need. Further, this combination would leave weaknesses with respect to non-linear mixed models,
reading complex raw data ¯les, reading old IBM style raw data ¯les. This combination also omits any of the
features for missing data that SAS has built-in that may not be available in the Stata user-written tools for
multiple imputation for missing data, and any of the methods for robust regression that SAS o®ers that is
not included with Stata.

谢谢版主，真的很全面