Stock Prediction Based on Long Short Term Memory

Recently, as the international situation becomes more complicated. The tendency of the stock is always with ups and downs since there are always sudden huge international focus to influence the routine trend of the stock. In this case, the robust of the Long Short-Term Memory model comes out to be significant. This essay is to discover the effect of Long Short-Term Memory prediction with certain noise by changing the noise sequence length. In addition, in the information explosion period, there are always layers to contain the input data of the feature. The essay is also to discuss if the feature is effective by grey relational analysis. Furthermore, in Long Short-Term Memory, multiple variables as well as single variable could be used to predict the price. The article is to describe the improvement of the model as well.