Multistage Design

Project Member

Emrah Kostem

The goal of this quarter

Design multi-stage association studies dynamically advancing one person at a time that approximates a random walk.

The Schedule for the quarter

  • Week 4: Decide on a project topic and create a page on the project wiki
  • Week 5: Read and study on random walks and understand multi-stage design better.
  • Week 6: Try to model one person advancement as a random walk.
  • Week 7: Get some sample data and convert the method into source code.
  • Week 8: Do some simulations comparing 2 stage design to random walk for power and cost.
  • Week 9: Prepare presentation and/or write-up of final project.
  • Week 10: Wrap-up and present project to class.

Project Description

In a different configuration where each individual has their own genotype we might be interested in collecting/selecting our association experiment data incrementally which could be modeled as a random walk. This would enable the study to select only the best candidates that results in a better power for the statistics.

Related Papers

April 25 2008

Wiki page created. Project started!

April 20 2008 - April 26 2008

  1. What I did this week: Selected a project topic, looked up some of the related literature, and created my wiki page.
  2. What I plan to do next week: Go through statistical material on random walks.
  3. How what I did compared to what I planned to do: Finishing the wiki page right now, so I got it all done for this week.

April 27 2008 - May 03 2008

  1. What I did this week: Started to read on chapters on sequential analysis from Siegmund [85].
  2. What I plan to do next week: Continue reading.
  3. How what I did compared to what I planned to do: I started reading on related literature,

May 04 2008 - May 10 2008

  1. What I did this week: Continued on reading, and did some computational experiments between fixed size and sequential analysis.
  2. What I plan to do next week: Formulate the problem to association studies.
  3. How what I did compared to what I planned to do: I continued reading on sequential analysis and did some comparisons.

May 11 2008 - May 17 2008

  1. What I did this week: Tried to formulate the problem as a binomial likelihood.
  2. What I plan to do next week: Incorporate composite hypothesis into the framework and prepare presentation.
  3. How what I did compared to what I planned to do: Formulated the problem in sequential analysis framework as a single hypothesis testing.

May 18 2008 - May 24 2008

  1. What I did this week: Prepared my presentation to the class.
  2. What I plan to do next week: Continue on composite hypothesis representation.
  3. How what I did compared to what I planned to do: I did my presentation.

May 25 2008 - May 31 2008

  1. What I did this week: Realized representing composite hypothesis is not trivial, rather it can be modeled through the random walk of the sum of the outcomes.
  2. What I plan to do next week: Go back and try to understand the proofs for Wald test boundaries.
  3. How what I did compared to what I planned to do: I stalled.

June 01 2008 - June 07 2008

  1. What I did this week: I couldn't be as productive as I wanted due to a take-home exam, but understood the proofs, Siegmund [85] has an example on composite hypothesis but it didn't click very much.
  2. What I plan to do next week: Come up with a representation of the likelihood for Alternate being not NULL hypothesis, and have the Type1 Type2 error probabilities in terms of them.
  3. How what I did compared to what I planned to do: I understood the proofs.

June 08 2008 - June 14 2008

  1. What I did this week: Had finals, and continued studying.
  2. What I plan to do next week: Quarter ended but this will be my summer project.
  3. How what I did compared to what I planned to do: I'm writing my project report.
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License