Data analysis tools

• Basic statistics
  • Averages and variability:
    x=randn(20,1);
mean(x)
median(x)
mode(x)
std(x)
var(x)
• T-tests
  • Paired:
    [h, p, ci, stats] = ttest(x – y) or ttest(x)
  • Unpaired:
    [h, p, ci, stats] = ttest2(x, y)
• ANOVA
  • One-way:
    [p, table, stats] = anova1(x)
    • Where x is a matrix with each column = a group
  • Two-way:
    [p, table, stats] = anova2(x)
    • Where x is a matrix with each columns = factor1 and rows = factor2
  • N-way:
    anovan.m
  • Repeated measures and mixed ANOVAs (not built in)
    RMAOV1.m, RMAOV2.m, RMAOV31.m, RMAOV32.m, RMAOV33.m, BWAOV2.m (between & within factors)
• Correlations and regression
  • [r p] = corr(x, y) (x and y must be column vectors)
    [b, bint, r, rint, stats] = regress(y, x) (y is a column vector and x is a matrix of column vectors = predictors)
    stepwise(x, y) produces an interactive figure for stepwise regression (x is a matrix of column vectors, y is a column vector)
    p = polyfit(x, y, n) where x and y are column vectors, n is the degree of polynomial (1=line); returns vector of coefficients corresponding to x^n, x^(n-1), …, x, 1.

Additional analysis tools

• Principal components analysis
  [coeff, score, latent, tsquare] = princomp(x)

Basic plotting tools

• Scatterplots and line plots:
  x = rand(10,1); y = rand(size(x));
plot(x, y);
plot(x, y, ‘g*-’, ‘linewidth’,4)
plot(x, y, ‘ro‘);
ye = .2*rand(size(x));
errorbar(x, y, ye, ‘.’)
• Bar graphs:
  bar(y)
• 3D plots:
  x = rand(10,1); y = rand(size(x)); z = rand(size(x));
plot3(x, y, z,’*’);
m = rand(5,5);
bar3(m)
• Adding text, titles, and labels:
  text(.3, .3, ‘Hello’, ‘fontsize’, 16);
xlabel(‘Whatever’);
ylabel(‘Something else’)
title(‘Anything’,’fontsize’,20)
• Modifying axis properties:
  errorbar(x, y, ye, ‘.’)
axis([-1 3 -1 3]);
axis equal
axis off
box off
set(gca, ‘xtick’, [.1:.2:30])
set(gca,’xticklabel’, [1:6:400])