|
wolffd@0
|
1 <html> <head>
|
|
wolffd@0
|
2 <title>Comparison of Matlab, R/S/Splus, Gauss, etc.</title>
|
|
wolffd@0
|
3 </head>
|
|
wolffd@0
|
4
|
|
wolffd@0
|
5 <body>
|
|
wolffd@0
|
6 <!--<body bgcolor="#FFFFFF"> -->
|
|
wolffd@0
|
7
|
|
wolffd@0
|
8 <h1>Comparison of Matlab, R/S/Splus, Gauss, etc.</h1>
|
|
wolffd@0
|
9
|
|
wolffd@0
|
10 <ul>
|
|
wolffd@0
|
11 <li> <a href="http://www.scientificweb.com/ncrunch/">Comparison of
|
|
wolffd@0
|
12 mathematical programs for data analysis</a>,
|
|
wolffd@0
|
13 Stefan Steinhaus, tech report, 2000.
|
|
wolffd@0
|
14 <br>
|
|
wolffd@0
|
15 This is a very detailed comparison of features and speed of several
|
|
wolffd@0
|
16 interactive scientific programming environments, e.g. Matlab,
|
|
wolffd@0
|
17 Mathematica, Splus.
|
|
wolffd@0
|
18
|
|
wolffd@0
|
19
|
|
wolffd@0
|
20 <!--
|
|
wolffd@0
|
21 <p>
|
|
wolffd@0
|
22 <li> <a href="Papers/gauss.econ.review.ps">Econometric
|
|
wolffd@0
|
23 programming environments: Gauss, Ox and S-PLUS</a>,
|
|
wolffd@0
|
24 Francisco Cribari-Neto.
|
|
wolffd@0
|
25 J. of Applied Econometrics, 12(1):77-89, 1997
|
|
wolffd@0
|
26 <br>
|
|
wolffd@0
|
27 Ox can not be used interactively, and has a C-style syntax (it even
|
|
wolffd@0
|
28 requires users to pre-declare variables!). Its only advantage is speed.
|
|
wolffd@0
|
29 S-Plus has tons of features and good documentation, but is slow.
|
|
wolffd@0
|
30 Gauss is somewhere in between.
|
|
wolffd@0
|
31
|
|
wolffd@0
|
32
|
|
wolffd@0
|
33 <p>
|
|
wolffd@0
|
34 <li> <a href="Papers/matlab.econ.review.ps">MATLAB as an econometric
|
|
wolffd@0
|
35 programming environment</a>,
|
|
wolffd@0
|
36 Francisco Cribari-Neto and Mark J. Jensen.
|
|
wolffd@0
|
37 J. of Applied Econometrics, 12(6):735-432, 1997.
|
|
wolffd@0
|
38 <br>
|
|
wolffd@0
|
39 The basic conclusion is that Matlab has excellent graphics and
|
|
wolffd@0
|
40 sparse-matrix facilities, but is slower than Gauss/Ox (especially on
|
|
wolffd@0
|
41 code
|
|
wolffd@0
|
42 with loops), and has few statistical routines built-in (one must buy the
|
|
wolffd@0
|
43 stats toolbox).
|
|
wolffd@0
|
44
|
|
wolffd@0
|
45
|
|
wolffd@0
|
46
|
|
wolffd@0
|
47 <p>
|
|
wolffd@0
|
48 <li> <a href="Papers/R.econ.review.ps">R: Yet another econometric
|
|
wolffd@0
|
49 programming environment</a>,
|
|
wolffd@0
|
50 Francisco Cribari-Neto and S. Zarkos.
|
|
wolffd@0
|
51 J. of Applied Econometrics, 14(3):319-329, 1999.
|
|
wolffd@0
|
52 <br>
|
|
wolffd@0
|
53 The basic conclusion is that R is much faster than Splus on
|
|
wolffd@0
|
54 code with loops, but a little bit slower on vectorized code. (Gauss/
|
|
wolffd@0
|
55 Ox is much faster than both; in my experience, R and Matlab have about
|
|
wolffd@0
|
56 the same speed.)
|
|
wolffd@0
|
57 However, R has much better memory management than Splus, and R is free. Otherwise, R/S/Splus
|
|
wolffd@0
|
58 are essentially the same.
|
|
wolffd@0
|
59 -->
|
|
wolffd@0
|
60
|
|
wolffd@0
|
61
|
|
wolffd@0
|
62 </ul>
|
