Long short-term memory networks for CSI300 volatility prediction with Baidu search volume

摘要

Intense volatility in financial markets affects humans worldwide. Therefore, relatively accurateprediction of volatility is critical. We suggest that massive data sources resulting from humaninteraction with the Internet may offer a new perspective on the behavior of market participantsin periods of large market movements. First, we select 28 key words, which are relatedto finance as indicators of the public mood and macroeconomic factors. Then, those 28 wordsof the daily search volume based on Baidu index are collected manually, from June 1, 2006 toOctober 29, 2017. We apply a Long Short-Term Memory neural network to forecast CSI300volatility using those search volume data. Compared to the benchmark GARCH model, our forecastis more accurate, which demonstrates the effectiveness of the LSTM neural network involatility forecasting.