2 A small tool to do the statistics legwork on benchmarks etc.
4 Prepare your data into two files, one number per line
6 ./ministat data_before data_after
10 You need at least three data points in each data set, but the more
11 you have the better your result generally gets.
13 Here are two typical outputs:
17 +--------------------------------------------------------------------------+
19 | |_________|______AM_______________|__A___________M_______________||
20 +--------------------------------------------------------------------------+
21 N Min Max Median Avg Stddev
22 x 5 36060 36138 36107 36105.6 31.165686
23 + 5 36084 36187 36163 36142.6 49.952978
24 No difference proven at 95.0% confidence
26 Here nothing can be concluded from the numbers. It _may_ be possible to
27 prove something if many more measurements are made, but with only five
28 measurements, nothing is proven.
33 +--------------------------------------------------------------------------+
37 | |_______________A_____M_________| |_M___A____| |
38 +--------------------------------------------------------------------------+
39 N Min Max Median Avg Stddev
40 x 5 0.133 0.137 0.136 0.1354 0.0015165751
41 + 5 0.139 0.14 0.139 0.1394 0.00054772256
42 Difference at 95.0% confidence
45 (Student's t, pooled s = 0.00114018)
47 Here we have a clearcut difference, not very big, but clear and unambiguous.