Price Prediction of Agricultural Products Based on Wavelet Analysis-LSTM

China is a big agricultural country; the fluctuation of agricultural products price impact people's financial life. Due to the natural climate, accidental event and other factors, the agricultural products price is unstable and changes quickly, it is difficult to predict agricultural products price. This paper takes Fuzhou cabbage as an example, and Wavelet Analysis (WA) is used to reduce noise of the cabbage data. And then normalize the data with the fine-tuned normalization. Finally, the normalized data are fed into Long Short-Term Memory (LSTM) model for prediction The new model named WA-LSTM based on both WA and LSTM can obtain better results than the classical LSTM model. The experiments show that this model achieved better performance and accuracy.