Table 3 The mean squared errors (MSE) and normalised mean squared errors (NMSE) of each model tested using leave-one-out cross-validation. The MSE and NMSE for the training and the validation sets, and the ratios of the MSE and NMSE of the validation set to that of the training set, respectively, are listed. The best-fitting model (Model: T_mean + R_tot) was compared with the alternative model (Model: T_mean + R_max), the fixed effects model (same predictors as the best-fitting model but without random effects) and the full model (mixed effects with all predictors).

From: The effects of seasonal climate variability on dengue annual incidence in Hong Kong: A modelling study