Interesting remarks about R back in 1999
Many people love Mathematica, but it's strength is symbolic mathematics, not data analysis. Googling for "data analysis in Mathematica and R" led me to an advertisement for a Mathematica add-on called "RLink", which is an "exciting new tool [to] leverage the statistical analysis power of R from within Mathematica! ... Overall, R contains perhaps the most advanced data analysis capabilities of any data analysis package ... . RLink ... allows Mathematica users to embed R in their Mathematica applications." (http://www.scienceops.com/Tools.asp?sID=215) In the late 1990s, colleagues raved about Mathematica and pushed me to use it. I tried several times, but never figured out how to get Mathematica to read a simple csv file and do things I did handily in S-Plus. Doubtless, that was partly my fault, because a book on "Data Analysis Using Mathematica" carries a copyright date of 1995. Some people had evidently solved the problems that overwhelmed my interest level at that time. Today, if I wanted symbolic mathematics, I might try Yacas (http://yacas.sourceforge.net/homepage.html -- and the Ryacas package). Mathematica is probably superior to Yacas, but I'd have to be convinced that the difference was sufficient to justify the extra expense. Just my 0.02 fraction of your favorite currency. Spencer
francogrex wrote:
Hi, this is not an R-help post, but I found this extract below that was written by a leading mathematician back in 1999 when he was talking about statistics and computing. I found it interesting to share and I ask your opinion do you think this still holds today or things have changed? Thanks. ?...we would also like to mention that in our opinion Mathematica provides an excellent and indeed unparalleled environment for mathematical statistical research. In comparison, no other computing environment provides such high quality capabilities simultaneously in: symbolics, numerics, graphics, typesetting and programming. Typically most researchers need to develop some code to implement their methods. Often the researcher?s code will only be executed a few times and the researcher?s main consideration is his time and effort as opposed to producing an cpu-efficient stand-alone software product. The importance of a powerful user-oriented programming language for researchers is sometimes lacking in other environment. ?comments have been made by research statisticians on the ease of programming in S and Splus as opposed to SAS. If the programming language is a natural extension of mathematical notation, this translates into ease and speed of development. This was found to be true in the past with APL, Splus and XLISP-STAT. We have found that Mathematica provides even more capability. However, for advanced state-of-the-art research and teaching in applied statistics and data analysis, R, Splus or XLISP-STAT may still be advantageous due to the wide usage by many leading researchers and the high quality functions for standard and advanced statistical methods that are available in the associated infrastructure (Statlib 1999). Furthermore, R and XLISPSTAT are freeware. However from the educational viewpoint, this advantage may not be so important since many students and researchers like to understand the principles involved. With Mathematica it is as easy to write out the necessary functions in Mathematica notation as it would be to explain the procedures in a traditional mathematical notation. In summary, Mathematica?s superior programming language is, in our opinion, one of its key strengths and advantages.?