The purpose of this lab is to carry out a power analysis and simulation study.
Your job this time is to provide a report for a meterologist collaborator (same as last time) providing power and sample size information under different scenarios for the clouds study. You can use the summary statistics from the clouds data to get you started with reasonable difference in means and group-specific SDs.
The report should have 2 parts: one part where the power calculations are computed from the t-distribution, and a second part where power is instead estimated by computer simulation. In each case you should compute sample sizes necessary to achieve 80%, 90% and 95% power at alpha = 0.01, 0.05 and 0.10 to detect differences for a few different combinations of difference and SD. Include scenarios where the groups are the same size and also where the groups are of different sizes.
You should describe to the meterologist collaborator what it is that you are trying to accomplish as well as the methods you are using. Results should be presented in tables, and might also be presented graphically.
The next class meeting is planned for 2 weeks from today, on 28 October 2009. Please feel free to email me in the meantime if you have any questions.
Your report can be in English or French. Please include .tex and R command files (or .Rnw file if you have used one) and .pdf file. Please follow the naming convention: surname2.tex, etc. (e.g. my files would be goldstein2.Rnw, goldstein2.pdf). For full re-write privileges, please email your report files to me (darlene.goldstein at epfl.ch) by Monday 1 November 2010.