One-factor ANOVA was performed with adjustment of error variances for individual genes so that it did not fall below the moving average error variance for genes with similar intensity, and with FDR method for assessing statistical significance [ 81].
They are image enhancement, digital filter design, adjustment of crossover errors, acoustic tomography, heat diffusion, curve fitting under a variety of norms, means of unit vectors, nonlinear curve fitting of spectral peaks, earthquake location, and vibrational eigenfrequency problem.
This resulted in the adjustment of standard errors and thereby adjustment of confidence intervals.
There was no evidence of autocorrelation of residuals (r=−0.16; p=0.40), so adjustment of standard errors was not deemed necessary.
There was no evidence of (first-order) autocorrelation of residuals (r=−0.0123, p=0.95), so no adjustment of standard errors was necessary.
Moreover, regression calibration requires post-fitting adjustment of standard errors, say by bootstrapping, which would be very burdensome computationally in the present setting.
This is because the MZ twins are supposed to share their genetic background which makes it somewhat uncertain whether the adjustment of standard errors, as described above, is correct for MZ twins.
Bonferroni adjustment of α-error rate was considered in case of multiple comparisons, in particular performing subgroup analysis.
An adjustment of the error probability would decrease the test power extremely so that the power of detecting existing mean differences would be very low.
When applying the Dunnett post hoc test, the number of significances was reduced by those endpoint pairs in which the t test showed a p value between 0.01 and 0.05 (significant, but not highly significant) according to the adjustment of the error rate for multiple comparisons.
*Prevalence ratio (PR) estimated by the Poisson regression model with robust adjustment of the standard errors.
