#################################################
Day 13: testing hypotheses with t and F contrasts
#################################################

5 / 22 / 2015

Reading:
    * slides from Jeanette Mumford about linear model. There are many slides, but see these between 17 and 77: 
      http://mumford.fmripower.org/2010_summercourse/stat_modeling_2010.pdf

    * Two recording commenting these slides - may (or may not) help you going through them:
	- slides 17-51: http://nipy.bic.berkeley.edu/practical_neuroimaging/glm_estimation.avi
	- slides 52-77: http://nipy.bic.berkeley.edu/practical_neuroimaging/hypothesistesting.avi

.. <!-- not this one until it is cleaner
    * Notebook on t / F - version of :
      http://nbviewer.ipython.org/urls/raw.github.com/practical-neuroimaging/pna-notebooks/master/GLM_t_F.ipynb --!>

Day:

* Simple model with intercept and slope
* Constructing your model
    - Motion parameters as confounds;
    - low frequencies as confounds;
* Estimating the parameters
* What is the distribution of the parameters?
    - bias
    - variance
* t and F tests, and associated contrasts, simple Anova;
* Example: Contrasts for motion;
* Correlation between regressors;