Journal: PLOS One
Article Title: Enhanced separation of long-term memory from short-term memory on top of LSTM: Neural network-based stock index forecasting
doi: 10.1371/journal.pone.0322737
Figure Lengend Snippet: Note: ( a – c ) forecasting results for the SZSE, the HSI, and the SSE, respectively. The blue lines represent the actual data trend, and the orange lines indicate the forecasted data trend; on the left side, RNN represents p ^ R N N , LSTM represents p ^ L S T M , RNN+LSTM represents p ^ R N N + ε ^ L S T M , and AR-RNN-LSTM represents the hybrid forecasting results. R represents the regression coefficient.
Article Snippet: Pawar, Jalem and Tiwari used an RNN to make forecasting on stock prices of Apple Inc., and the forecasting accuracy exceeded 95%, with a loss close to 0.1% [ ].
Techniques: