Table 3.

Model validation metrics for four specifications of models of Lyme disease incidence (see Methods: Model validation). The model validation metrics shown are the root-mean-square error (RMSE) and correlation coefficient (r) for estimated versus observed Lyme disease incidence in the testing data sets. The observed and estimated average (± 1 standard deviation) annual Lyme disease incidence is also shown for each region and each model specification. Model validation was performed using data from 2008 – 2017 (the years with complete data for all predictors).

		Main Model			Model Spec. 2			Model Spec. 3			Model Spec. 4
	Observed annual incidence	Est. annual inc.	RMSE	r	Est. annual inc.	RMSE	r	Est. annual inc.	RMSE	r	Est. annual inc.	RMSE	r
NE	48.9 ± 17.4	51.3 ± 13.3	38.970	0.853	51.8 ± 15.6	39.138	0.851	49.4 ± 9.4	65.419	0.458	51.2 ± 13.2	38.343	0.858
MW	14.5 ± 3.2	12.7 ± 2.1	15.709	0.903	12.6 ± 3.1	15.706	0.902	14.2 ± 4.0	29.023	0.602	12.7 ± 2.1	15.49	0.906
PC	0.8 ± 0.3	0.8 ± 0.1	1.739	0.402	0.9 ± 0.3	1.739	0.404	0.9 ± 0.1	1.777	0.282	0.8 ± 0.1	1.736	0.423
PS	0.9 ± 0.6	0.8 ± 0.4	1.682	0.264	0.8 ± 0.4	1.682	0.268	0.8 ± 0.2	1.316	0.321	0.8 ± 0.4	1.747	0.262
SW	0.4 ± 0.3	0.4 ± 0.2	5.169	0.071	0.4 ± 0.3	5.170	0.070	0.3 ± 0.2	5.131	0.040	0.4 ± 0.2	5.157	0.086
SE	0.5 ± 0.2	0.5 ± 0.2	1.685	0.323	0.5 ± 0.2	1.694	0.313	0.5 ± 0.2	1.725	0.172	0.5 ± 0.2	1.682	0.326