Exploring the predictability of range-based volatility estimators using\n RNNs

2018/03/19 by Gábor Petneházi, Petneházi, Gábor, József Gáll +1
Decision Sciences · Economics, Econometrics and Finance · #Computational Finance (q-fin.CP) #FOS: Computer and information sciences #FOS: Economics and business #Financial Risk and Volatility Modeling #Forecasting Techniques and Applications #Machine Learning (stat.ML) #Statistical Finance (q-fin.ST) #Stock Market Forecasting Methods

paper · pdf · doi:10.48550/arxiv.1803.07152

openalex publication_date 2018/03/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We investigate the predictability of several range-based stock volatility\nestimators, and compare them to the standard close-to-close estimator which is\nmost commonly acknowledged as the volatility. The patterns of volatility\nchanges are analyzed using LSTM recurrent neural networks, which are a state of\nthe art method of sequence learning. We implement the analysis on all current\nconstituents of the Dow Jones Industrial Average index, and report averaged\nevaluation results. We find that changes in the values of range-based\nestimators are more predictable than that of the estimator using daily closing\nvalues only.\n

Related

• Recurrent Neural Networks with more flexible memory: better predictions than rough volatility
• Can LSTM outperform volatility-econometric models?
• Modeling and Forecasting Stock Market Volatility Using Recurrent Deep Learning Architectures: Evidence from RNN, LSTM, and GRU Models
• Volatility Forecasting with 1-dimensional CNNs via transfer learning
• Benchmarking Deep Sequential Models on Volatility Predictions for Financial Time Series