CogMaster Statistics course: Python materials

Download

• HTML and example files
• Source code (github)

Python materials for the statistics course of the CogMaster.

Authors: S. Charron, G. Varoquaux

Tip

Why Python for statistics in experimental psychology?

R is a language dedicated to statistics. Python is a general purpose language with statistics module. R has more statistical analysis features than Python, and specialized syntaxes. However, when it comes to building complex analysis pipelines that mix statistics with e.g. image analysis, as in the case of brain imaging, the richness of Python is an invaluable asset.

See also

• MNE Python: Analysis of EEG and MEG in Python
• Nipy The NeuroImaging in Python ecosystem
• PyschoPy: presention of stimuli and collection of data for experimental psychology

See also

Scipy lecture notes

The scipy lecture notes have a chapter on statistics in Python that is kept up to date and is a good complement to these notes for statistic topics not specific to experimental psyschology.

Note

To install Python, we recommend that you download Anaconda Python

---

• 1. Basic statistics
  • 1.1. Interacting with data
    • 1.1.1. Work environment: IPython
    • 1.1.2. Basic array manipulation: numpy
    • 1.1.3. Basic plotting: pylab
      • 1.1.3.1. The box plot
      • 1.1.3.2. More plots
    • 1.1.4. Mixed-type data: pandas
      • 1.1.4.1. Inputing data
      • 1.1.4.2. Manipulating data
      • 1.1.4.3. Plotting data
  • 1.2. Hypothesis testing: two-group comparisons
    • 1.2.1. Student’s t-test
      • 1.2.1.1. 1-sample t-test
      • 1.2.1.2. 2-sample t-test
    • 1.2.2. Paired tests
  • 1.3. Linear models, ANOVA
    • 1.3.1. A simple linear regression
    • 1.3.2. Multiple Regression
    • 1.3.3. Post-hoc ANOVA: contrast vectors
• 2. FMRI signals
  • 2.1. Getting ready: installing the software
    • 2.1.1. Installation
      • 2.1.1.1. Sanity check
      • 2.1.1.2. Download the neurodebian virtual machine
      • 2.1.1.3. Install VirtualBox
      • 2.1.1.4. Configure the neurodebian virtual machine
  • 2.2. Activation IRMf
  • 2.3. BOLD signal
  • 2.4. fMRI stimulation data
  • 2.5. Convolution
  • 2.6. fMRI scanner data
  • 2.7. Analysis principle
  • 2.8. Analysis example
  • 2.9. Contrasts for T- and F-tests
  • 2.10. Another example
  • 2.11. Multiple comparisons
  • 2.12. Second-level analysis
• 3. Examples
  • 3.1. This course’s code examples
  • 3.2. Matplotlib demo examples
  • 3.3. Solutions to the exercises of the course