Really depends on what you want to do and different area.
Generally,
1. In industry, always C/C++, it's like a definition, I got a bookedited by MichaelPage for QF. Most investment bank, financialconsultancy perfer C++. As to Maple, Mathematica, Minitab, Fortran, allthese belong to academia, it's useful when you publish paper, makepresentation, holding workshop.
2. Regarding statistical software, SAS and S-Plus are mostpopulars.Both academia and industry. Nor to say SAS, just S-Plus, isdeveloped by Insight, which is a big financial software provider. Youmay find why it's popular from this point. As to R, definetely no oneuse it in the industry...I am not sure about the academic.
3. The most important is to pick one up, which is particulay fit foryour research or work. Then it will become good one, and u will becomethe expert on that.
Very good example is that every year, our rearch center is holdinig aworkshop called Optimisation, on that workshop we introduce AMPL, whichis very good in stochastic programming and linear programming, some ofour presenter still perfer MPL, or SPInE.......
For more details you may visit my webpage @
http://mam3xs.blogspot.com
I am not sure if it could be open in China, or that you need to go t http://finmath.blogchina.com
hope they are useful
EView: strong on time series analysis and easy to use with good graphics.
Shazam: up to date with many tests.
Limdep: strong on discrete choice, limited dependent variables, non-linear regressions and complex hypothesis tests, but less for time series analysis.
Stata: strong all around.
SAS, SPSS: do not have the tests needed for econometrics.
STATA is my favoriate. powerful enough, but so much more user-friendly than SAS. many believe that someday SAS will be replaced by STATA. SAS is still used in many places including govt and probably industries (not because it is so much better but because STATA just started in late 1980s). STATA is quickly becoming the most popular econometrics software in business schools (you know they teach the next generation financial analysts and managers). I've heard that S is also good. SPSS is still the favoriate for most sociologists and psychologists for its ease of use. check a UCLA website for comparison.