Skip to main content
Energy Informatics

Table 3 Metrics for the models based on portfolio level

From: Probabilistic forecast of electric vehicle charging demand: analysis of different aggregation levels and energy procurement

Model

nRMSE*

MASE*

R2**

PS low Q***

PS high Q***

IS***

LinR

0.883

0.777

0.788

–

–

–

Bagging

0.432

0.358

0.949

1.815

2.815

133.882

GradientB

0.508

0.498

0.930

3.526

4.156

203.921

Ada

0.418

0.355

0.952

1.866

2.767

147.959

Random Forest

0.411

0.374

0.954

2.843

4.474

95.621

LSTM

0.610

0.603

0.899

37.899

76.411

150.056

CNN

0.630

0.538

0.892

42.207

71.183

140.144

NN

0.609

0.627

0.899

115.093

95.773

247.339

  1. Best result bold, second underlined
  2. *Unitless, **in %, ***in W
Back to article page