Research Article
Evaluating the Performance of Feature Selection Methods Using Huge Big Data: A Monte Carlo Simulation Approach
Table 1
Variable selection under multicollinearity from Monte Carlo Simulation.
| | Models | ∑ = 0.25, P = 50 | ∑ = 0.25, P = 70 | | n = 80/160/320 | Potency | Gauge | Potency | Gauge |
| | MCP | 1/1/1 | 0.04/0.02/0.02 | 0.99/1/1 | 0.05/0.02/0.01 | | E-SCAD | 1/1/1 | 0.12/0.10/0.10 | 1/1/1 | 0.11/0.10/0.09 | | AEnet | 0.99/1/1 | 0.01/0/0 | 0.99/1/1 | 0.02/0/0 | | Autometrics | 0.99/1/1 | 0.04/0.01/0.01 | 0.99/1/1 | 0.04/0.01/0.01 | | n = 80/160/320 | ∑ = 0.50, P = 50 | ∑ = 0.50, P = 70 | | MCP | 0.99/1/1 | 0.06/0.02/0.01 | 0.99/1/1 | 0.09/0.01/0.01 | | E-SCAD | 1/1/1 | 0.10/0.07/0.06 | 0.99/1/1 | 0.09/0.06/0.06 | | AEnet | 0.99/1/1 | 0/0/0 | 0.99/1/1 | 0/0/0 | | Autometrics | 0.99/1/1 | 0.02/0.01/0.01 | 0.98/1/1 | 0.06/0.01/0.01 | | n = 80/160/320 | ∑ = 0.90, P = 50 | ∑ = 0.90, P = 70 | | MCP | 0.68/0.94/0.99 | 0.19/0.22/0.09 | 0.59/0.92/0.99 | 0.16/0.23/0.09 | | E-SCAD | 0.91/0.98/0.99 | 0.13/0.09/0.03 | 0.89/0.98/0.99 | 0.12/0.09/0.03 | | AEnet | 0.93/0.98/0.99 | 0/0/0 | 0.91/0.98/0.99 | 0/0/0 | | Autometrics | 0.63/0.89/0.99 | 0.06/0.02/0.02 | 0.61/0.87/0.99 | 0.17/0.03/0.01 |
|
|
