Table 2 Three types of multivariate regression models.

From: Association of meteorological factors with the frequency of primary rhegmatogenous retinal detachment in Japan

Model

Variables

AICc

BIC

VE

CV-VE (95% CI)

Model A

All variables (n = 36)

472.37

422.05

96.03%

 − 73.86% (− 625.61 to 83.96%)

Model B

Variables with P < 0.05 in univariate analysis (n = 8)

337.38

352.63

89.05%

83.76% (70.72 to 93.16%)

Model C

Correlated variables in Model B were removed (n = 6)

331.97

344.48

89.04%

84.82% (72.18 to 93.72%)

  1. AICc Akaike’s information criterion with correction for small sample size, BIC Bayesian information criterion, CI confidence interval, CV-VE proportion of variance explained by cross-validation method, VE proportion of variance explained.
Back to article page