Online Stock Price Prediction Based on Interval Data Analysis

The continuous increase in per capita income makes more residents choose stocks as a new investment method, so how to more accurately judge their price trends has become increasingly important. In most traditional time series analyses, models are built on basis of closing price, from the perspective of probability. This paper introduces the interval data into the stock price prediction task and proposes an attention mechanism-based long short-term memory (LSTM) model. Specifically, borrowing the idea from the sequence-to-sequence (seq2seq) model, the LSTM is first used as an encoder to encode the input sequence. Then the attention mechanism is used to capture the most useful information for the current output based on the encoded features. Finally, another LSTM model is used as a decoder to decode the encoded data features and obtain the prediction results. Experimental results show that the proposed model significantly improves the prediction accuracy.