Skip to main content

Table 3 Predicting criteria of the ZIP and ZINB models with the validation dataset

From: Using a zero-inflated model to assess gene flow risk and coexistence of Brassica napus L. and Brassica rapa L. on a field scale in Taiwan

Dispersal kernel functionZINB model
RMSEAdj. R2AICBICMSPR
Log-sech0.00830.34657− 7650.3− 7644.90.000068767
Exponential power0.00830.34681− 7650.5− 7645.20.000068742
Power law0.008330.34179− 7644.4− 7639.10.000069270
Logistic0.008310.34446− 7647.7− 7642.30.000068989
2Dt0.00830.34595− 7649.5− 7644.10.000068832
Gamma0.00830.3466− 7650.3− 7644.90.000068764
WALD0.008310.34404− 7647.2− 7641.80.000069033
Weibull0.00830.3466− 7650.3− 7644.90.000068764
Neg. Exponential0.008360.33831− 7641.2− 7633.80.000069723
Log-normal0.00830.3456− 7649.1− 7643.70.000068869
Gaussian0.008320.34363− 7647.7− 7640.30.000069163
  1. RMSE: root mean square error; adj. R2: adjusted coefficient of determination; AIC: Akaike’s information criterion; BIC: Schwarz’s Bayesian information criterion; MSPR: mean squared prediction error