Table 1 Proposed and implemented models. NN denotes that the parameters are not needed for a model. Alpha is the parameter α and is needed only in generalized Poisson regression models. UPDN_1 denotes the univariate Poisson deep neural network with 1 hidden layer, UPDN_2 denotes a UPDN with two hidden layers and so on. MPDN_1 denotes the multivariate Poisson deep neural network with 1hidden layer, MPDN_2 denotes a MPDN with two hidden layers, and so on
ModelModel nameAbbreviation of modelHidden layerAlpha
1Univariate Poisson deep neural networkUPDN_11NN
2Univariate Poisson deep neural networkUPDN_22NN
3Univariate Poisson deep neural networkUPDN_33NN
4Univariate Poisson deep neural networkUPDN_44NN
5Multivariate Poisson deep neural networkMPDN_11NN
6Multivariate Poisson deep neural networkMPDN_22NN
7Multivariate Poisson deep neural networkMPDN_33NN
8Multivariate Poisson deep neural networkMPDN_44NN
9Univariate Generalized Poisson Elastic net regressionGPR_0.7500.75
10Univariate Generalized Poisson Elastic net regressionGPR_0.500.5
11Univariate Generalized Poisson Elastic net regressionGPR_0.2500.25
12Univariate Generalized Poisson Lasso regressionGPR_Lasso01
13Univariate Generalized Poisson Ridge regressionGPR_Ridge00