Statistical Methods for Life and Health Sciences
- (Project_2_1) You should collect a raw data set from your field of study/research. Follow the guidelines we have discussed in class to make sure your design is appropriate (e.g., controlled, randomized, blinded, large-sample-size(s), etc.). Identify a parameter of interest for the population you are studying (e.g., population mean) and construct a data-driven estimate of a 95% confidence interval for this parameter. What is an unbiased estimate of the parameter of interest? What is its precision? [In a case you do not have a ready access to a proper dataset, here I have several datasets you may consider for this project. I do recommend you first try to dig out data from your specific area of interest, though. Talk to colleagues, mentors, established investigators, etc.] Can you identify a distribution, from the ones we presented in class, that is a reasonable fit for your data? Perhaps, there is a distribution we did not identify in class, which other people have labeled and used in other situations that may better describe your data. What is it?
\Ivo D. Dinov, Ph.D., Departments of Statistics and Neurology, UCLA School of Medicine/