\(\newcommand{L}[1]{\| #1 \|}\newcommand{VL}[1]{\L{ \vec{#1} }}\newcommand{R}[1]{\operatorname{Re}\,(#1)}\newcommand{I}[1]{\operatorname{Im}\, (#1)}\)

Exploring the general linear model

Python path

• Using PYTHONPATH;

• stimuli.py file;

• test_stimuli.py file;

• mkdir code
  mv stimuli.py code
  

• Install pytest if you haven’t got it already:

  pip install pytest
  

• Show that the tests don’t work yet:

  py.test test_stimuli.py
  

• Set Python path;

• Finally:

  mv test_stimuli.py code
  py.test code/test_stimuli.py
  

Simple and multiple regression

• finish going through the introduction to the General Linear Model;

• we get the same results in R:

  psychopathy = c(11.416,   4.514,  12.204,  14.835,
                  8.416,   6.563,  17.343, 13.02,
                  15.19 ,  11.902,  22.721,  22.324)
  clammy = c(0.389,  0.2  ,  0.241,  0.463,
             4.585,  1.097,  1.642,  4.972,
             7.957,  5.585,  5.527,  6.964)
  res = lm(psychopathy ~ clammy)
  print(summary(res))
  

• Inverse of a diagonal matrix;

• Subtracting the mean from a vector;

• Least-squares regression exercise;

• on matrix rank;

• A worked example of the general linear model.

Correlation again

• Make sure you have stimuli.py and pearson.py on your Python path. (pearson.py comes from the exercise in Pearson 1D and 2D function exercise);
• Correlation per voxel, in 2D.

Reading and homework for next week

• Finish the Least-squares regression exercise – see GLM exercise, with some github practice;
• Do preliminary work on projects to prepare for project pitch next week.