I'm a PhD student in theoretical computer science looking for a quick primer on basic statistics, such as what you might find in an undergrad stats curriculum at a top university. I have a solid math background and expect to cover the material in an hour or two. This would be done over google hangout/chat or equivalent.
Things I would like to learn:
-- What the common distributions are (normal, log normal, poisson, cauchy, chi-squared, .. ?) along with their basic properties.
-- Frequentist inference, eg. how to verify or reject a hypothesis with a certain p-value, do logistic regression, calculate marginals, etc
-- How to design experiments, and calculate power given a sample size.
-- Anything else you think I should know for running basic experiments on large data sets.
I think I understand the bayesian side of things well enough, and don't need help there.
Please start your application with an answer to the following question:
Bob has some hypothesis H, and he is going to collect samples until he can accept or reject H with p < .01. What is wrong with this approach?
I expect to pay $60/hr spent teaching, with the understanding that this will probably take some (otherwise uncompensated) prep time if you don't regularly teach statistics to people.