Skip to main content

Table 4 Relative errors of prediction models

From: Characterizing financial markets from the event driven perspective

Model W MAE MSE
Decision tree cent 0.957 0.968
  clst 1.002 0.984
  cos 0.929 0.972
  count 1.013 1.008
Gaussian process cent 0.992 0.996
  clst 1.001 1.002
  cos 0.994 1.002
  count 0.990 1.011
KernelRidge cent 2.034 2.331
  clst 4.053 6.123
  cos 2.032 3.670
  count 4.431 10.695
Linear regression cent 5.298 13.083
  clst 6.750 18.585
  cos 5.538 20.137
  count 5.374 23.878
Nearest neighbors cent 0.936 0.947
  clst 0.971 0.983
  cos 0.934 0.977
  count 1.010 0.984
Neural net cent 0.463 0.164
  clst 0.499 0.180
  cos 0.505 0.199
  count 0.636 0.285
Random Forest cent 0.951 0.974
  clst 0.968 0.936
  cos 0.963 1.008
  count 1.004 0.993
  1. This table presents relative mean absolute error (MAE) and mean squared error (MSE) of various models trained on combined financial time series and network obtained from the news. The baseline for relative errors calculation is in each case the same model trained solely on financial data set. Bold number show which model had the largest decrease in error when information from news was included in the data set