IntelliPortfolio: Intelligent Portfolio for Enhanced Index Tracking Using Clustering and LSTM

Table 4

Hyperparameters for EIT algorithms.

Algorithms	Hyperparameters
GA-RRL	SequenceStep:30; WindowLen:20; LearningRate: 0.0001; stddev: 0.05; ElitistsNum: 10; ChoosenEliNum: 4; TrainEpoch: 15; TradeEpoch: 10; PopulationSize: 50; GenerationNum: 20; CrossoverProb: 0.5; MutationProb: 0.2
DDPG	: 0.3; MaxEpisodes: 1000; MaxEpSteps: 200; LrA: 0.001; LrC: 0.002; : 0.9; : 0.1; MemoryCapacity: 1000; BatchSize: 32
RRL	LearningRate: 0.0001
DPG	TrainingSteps: 8000; LearningRate: 0.0001; BatchSize: 40; FeatureNum: 3
HGA	Epsilon: 0.01; delta: 0.5; WinLen: 300; gamma: 0.1; Lambda: 0.3, : 2.0; PopuationSize: 100; LoopNum: 10000; CrossoverProb: 0.5; MutationProb: 0.5