From: Predicting LQ45 financial sector indices using RNN-LSTM
Authors (Ref) | Case study | Method | MAPE |
---|---|---|---|
Patel et al. [25] | CNX Nifty and S&P Bombay Stock Exchange (BSE) Sensex | ANN SVR-ANN SVR SVR-SVR RF SVR-RF | 2.96 (average) 2.62 (average) 2.69 (average) 2.63 (average) 3.02 (average) 2.70 (average) |
Hiransha et al. [26] | MARUTI, HCL, AXIS BANK, BANK OF AMERICA, CHESAPEAK ENERGY | RNN LSTM CNN MLP | 8.00 (average) 7.29 (average) 6.84 (average) 6.89 (average) |
Sun et al. [27] | S&P 500, SHCI | ARIMA MLPNN SVR ELM LSTM | 5.07 (average) 2.89 (average) 2.08 (average) 1.50 (average) 0.99 (average) |
Jin et al. [9] | AAPL (Apple stock) | LSTM LS_RF S_LSTM S_AM_LSTM S_EMDAM_LSTM | 4.58 3.15 2.23 1.82 1.65 |
Nabipour et al. [28] | Diversified financials from the Tehran Securities Exchange Technology Management Co (TSETMC) | Decision tree Bagging Random forest Adaboost Gradient boosting XGBoost ANN RNN LSTM | 2.07 1.91 1.91 1.59 1.70 1.72 3.86 1.85 0.60 |
Our approach | LQ45 financial indices from IDX | 3 layers LSTM | 0.26 (average) |