Matlab's strengths for statistics are its computing power and its programming language.
The Statistics Toolbox does many things that basic Matlab does not do. Try
>>help stats
to see the commands. Matlab can be run on godzilla.acpub.duke.edu or craven2.acpub.duke.edu (usually faster), or any other installation. Its free counterpart Octave is not bad but lacks some essential statistical functions in the Toolbox. The Matlab functions can be examined with "type", but many of them will not run in Octave. The free statistics software R has some advantages, but we will only spend time in class on Matlab.
Selected Matlab commands:
help, quit, diary, !, who, whos, more on,
load, tdfread, textread, save, reshape,
median, mean, std, prctile (what formula do they use?), format,
hist, histfit, boxplot, corrcoef, cdfplot, legend,
plot, polyfit, regress, lsline, hold,
crosstab, diff, cumsum, ./
Getting tabular data into matlab is a bit harder than with some other
statistics software. "tdfread" is pretty good, but for a space-separated
file you need to specify the delimiter--tdfread('data.txt', ' ') (similarly
for a comma).
It also expects a line of column labels at the top, and it will take
the first line of data and use it as labels if there are no labels! An input
option for numeric data without column labels is "load".