Mar 3 Lecture - My Results 50 Replicate Simulations...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
1 My Results – 50 Replicate Simulations ± Remember, your results may vary due to the randomness introduced by the simulation ± What % of the time did I get p< 0.008469 by chance (what is the empiric p-value)? ± There is a lot of variability here – the largest estimate is over twice the size of the smallest! 16% 8% 6% 16% 12% My Results – 250 Replicate Simulations ± What % of the time did I get p< 0.008469 by chance (what is the empiric p-value)? ± Much less variability now, but still the largest result is 29% larger then the smallest result ± More replicates lower run-to-run variation ± A bigger simulation would be more accurate 14.8% 14.8% 12.4% 16.0% 14.4%
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 Are the Unknown Results Significant? ± Despite the run-to-run variability, all of the simulations indicate that the original results are not significant ± It is relatively easy (about 14% chance) to get a p-value of 0.008469 by chance when there is no association ± But something seems wrong with our simulation – what? ± Remember that we used 10 markers in our study Correction for Independent Tests ± If we had conducted 10 independent tests, we could just multiply our nominal p-value by 10 to get a Bonferroni corrected p-value ± 0.008469 x 10 = 0.08469 ± Yet we know that these SNPs are not independent, so our empiric p-value should be smaller then the Bonferroni corrected one ± But our simulation gives an empiric p-value of about 0.14 – so the results don’t make sense!
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 04/04/2011 for the course GENETICS 302 taught by Professor Hey during the Spring '11 term at Rutgers.

Page1 / 8

Mar 3 Lecture - My Results 50 Replicate Simulations...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online