An LSTM based classification method for time series trend forecasting
Yuqi Liu, Zhongfeng Su, Hang Li, Yulai Zhang · 2019
Trend forecast of time series is an essential task in many fields. Deep neural network with recurrent structures is developed recently to extract information from sequential data. LSTM is a special recurrent neural network that learns long term dependencies. It is suitable for predicting time series with both long term and short term dependencies. In this paper, we implement this model on time series trend forecasting problems. The results outperform that of the conventional auto regression models and also significantly higher than the random guess results on the stock data sets, which are very close to random walk sequences.