Skip to main content
. Author manuscript; available in PMC: 2022 Feb 16.
Published in final edited form as: Stat Methods Med Res. 2019 Jun 7;29(4):1167–1180. doi: 10.1177/0962280219852720

Figure 1.

Figure 1.

Simulation results for mSRI and CMI under the null of no SNP-expression association for a range of expression-trait effect sizes. Type-1 error is estimated as the proportion of simulations for which the test statistic H0: γ = 0 is significant at the 0.05 level. Coverage is the proportion of 95% confidence intervals that cover the true parameter value used for the simulation – under the complete null, coverage is the same as one minus the type-1 error rate. Estimation error is defined as the difference between the estimated parameter for association between expression and the trait and the true value used for the simulation (=0 in this example). (a) Power (type-1 error rate) and coverage. (b) Absolute deviation (median and IQR).