|
With the vigorous development of Taiwan’s economy, in order to diversify the risks of investment and financial management, investors often invest in many projects in order to obtain maximum benefits, and there are high returns and high risks in stock investment, so in order to allow investors to have a basis and reduce risks For the purpose of, we will use the historical data of the stock market to analyze and predict the rise and fall.
This study uses decision trees, logistic regression and LSTM to predict the rise and fall of the three ETF stocks of Taiwan 50 (0050), Fubon Technology (0052) and Yuanta Financial (0055), and adds the Average Directional Indicator (Average Directional Indicator). , ADX), Stochastic Oscillator (KD), Momentum (MOM), Relative Strength Index (RSI), Moving Average Convergnece & Divergence (MACD) comparison, its The results show that the data with technical indicators has a better prediction effect, and LSTM has the highest accuracy. Then, we added the index data to predict the exact path of stocks on the 5th, 10th, and 20th. The results show that the accuracy of LSTM is still the highest regardless of the length of the forecast.
|