Research Article

A Hybrid Prediction Method for Stock Price Using LSTM and Ensemble EMD

Table 1

Prediction results of time series.

MethodSinSP500HSIDAXASX

SVRRMSE0.2824091219.9047831334.9912150.6534530.206921
MAE0.238586990.6390201282.8624980.4157960.146022
R20.968293−1.665447−1.0887310.9382620.959223

BARDRRMSE0.012349114.140660326.4871070.3870250.110975
MAE0.011054110.146472233.6419800.2561250.075355
R20.9999390.9982750.9922130.9783430.988271

KNRRMSE0.5678071198.484251553.6419430.7407630.303352
MAE0.443162937.309021452.3002930.4452120.186848
R20.871824−1.572662−1.2395160.9206610.912361

LSTMRMSE0.05959730.261563306.4815490.6546620.135236
MAE0.05137619.164419224.2500310.5046450.106639
R20.9985620.9971280.9802700.9627500.931165

Proposed LSTM-EMDRMSE0.03247876.56208278.4879940.4039720.066237
MAE0.02617547.31700971.0628340.2177330.045643
R20.9995690.9855230.9875940.9794440.996174

Proposed LSTM-EEMDRMSE0.20135848.87833148.7839060.3243030.101511
MAE0.16434838.55650038.5648120.2477220.081644
R20.9829800.9941710.9884510.9868320.991071