Machine Learning for Predicting Stock Return Volatility

Filipović, Damir; Khalilzadeh, Amir

doi:10.2139/ssrn.3995529

Cited by 6 publications

(5 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the efficacy of machine learning methods in our study adds to the insights provided by researchers such as Filipovic &Khalilzadeh (2021) andD'Ecclesia &Clementi (2021), the world of cryptocurrency and innovations like those explored by Zahid et al (2022) remain outside our scope. These variations underscore the complexity and dynamism of the field, where machine learning methods are playing an increasingly prominent and nuanced role.…”

Section: Discussionmentioning

confidence: 99%

“…However, the superiority is not uniform; Shen et al (2021) and Christensen et al (2022) observed that machine learning methods do not always efficiently capture extreme market events. On the other hand, several authors have emphasized the efficacy of these methods, such as Kristjanpoller et al (2014), Filipovic &Khalilzadeh (2021), andD'Ecclesia &Clementi (2021), who stressed the ability of neural networks to uncover complexities in the volatility of stock returns.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

A Latin American Market Asset Volatility Analysis: A Comparison of Garch Model, Artificial Neural Networks and Support Vector Regression

CHUNG,

ESPINOZA

2023

Appl. Comput. Sci.

View full text Add to dashboard Cite

The objective of this research was to compare the effectiveness of the GARCH method with machine learning techniques in predicting asset volatility in the main Latin American markets. The daily squared return was utilized as a volatility indicator, and the accuracy of the predictions was assessed using root mean square error (RMSE) and mean absolute error (MAE) metrics. The findings consistently demonstrated that the linear SVR-GARCH models outperformed other approaches, exhibiting the lowest MAE and MSE values across various assets in the test sample. Specifically, the SVRGARCH RBF model achieved the most accurate results for the IPC asset. It was observed that GARCH models tended to produce higher volatility forecasts during periods of heightened volatility due to their responsiveness to significant past changes. Consequently, this led to larger squared prediction errors for GARCH models compared to SVR models. This suggests that incorporating machine learning techniques can provide improved volatility forecasting capabilities compared to the traditional GARCH models.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

A Latin American Market Asset Volatility Analysis: A Comparison of Garch Model, Artificial Neural Networks and Support Vector Regression

CHUNG,

ESPINOZA

2023

Appl. Comput. Sci.

View full text Add to dashboard Cite

show abstract

“…As can be expected, rough volatility models outperform GARCH-based models for volatility prediction. Using LSTMs for volatility prediction is demonstrated, e.g., in Filipović and Khalilzadeh (2021); Ganesh and Rakheja (2018); Kim and Won (2018); Rosenbaum and Zhang (2022), which use various types of predictors (including GARCH models) and architectures. Notably, Rosenbaum and Zhang (2022) show that the average predictions of 10 stacked LSTMs with the past volatility and price return as predictors match the performance of rough volatility and has universality properties; i.e., a single model is able to predict the volatility dynamics of many assets.…”

Section: Volatility Predictionmentioning

confidence: 99%

Multi-Timescale Recurrent Neural Networks Beat Rough Volatility for Intraday Volatility Prediction

Challet,

Ragel

2024

Risks

View full text Add to dashboard Cite

We extend recurrent neural networks to include several flexible timescales for each dimension of their output, which mechanically improves their abilities to account for processes with long memory or highly disparate timescales. We compare the ability of vanilla and extended long short-term memory networks (LSTMs) to predict the intraday volatility of a collection of equity indices known to have a long memory. Generally, the number of epochs needed to train the extended LSTMs is divided by about two, while the variation in validation and test losses among models with the same hyperparameters is much smaller. We also show that the single model with the smallest validation loss systemically outperforms rough volatility predictions for the average intraday volatility of equity indices by about 20% when trained and tested on a dataset with multiple time series.

show abstract

“…Instead of predicting stock returns, ML can also be used to predict other relevant metrics, such as market betas (Jourovski et al, 2020) or stock volatilities (Filipovic and Khalilzadeh, 2021). However, the same caveat that applies to enhancing factors applies here as well, namely that simply incorporating some information from other variables such as value and momentum can already lead to better risk forecasts.…”

Section: Predicting Metrics Other Than Returnmentioning

confidence: 99%

How Can Machine Learning Advance Quantitative Asset Management?

Blitz¹,

Hoogteijling²,

Lohre³

et al. 2023

JPM

View full text Add to dashboard Cite

The emerging literature suggests that machine learning (ML) is beneficial in many asset pricing applications because of its ability to detect and exploit nonlinearities and interaction effects that tend to go unnoticed with simpler modelling approaches. In this paper, we discuss the promises and pitfalls of applying machine learning to asset management, by reviewing the existing ML literature from the perspective of a prudent practitioner. The focus is on the methodological design choices that can critically affect predictive outcomes and on an evaluation of the frequent claim that ML gives spectacular performance improvements. In light of the practical considerations, the apparent advantage of ML is reduced, but still likely to make a difference for investors who adhere to a sound research protocol to navigate the intrinsic pitfalls of ML.

show abstract

Machine Learning for Predicting Stock Return Volatility

Cited by 6 publications

References 33 publications

A Latin American Market Asset Volatility Analysis: A Comparison of Garch Model, Artificial Neural Networks and Support Vector Regression

A Latin American Market Asset Volatility Analysis: A Comparison of Garch Model, Artificial Neural Networks and Support Vector Regression

Multi-Timescale Recurrent Neural Networks Beat Rough Volatility for Intraday Volatility Prediction

How Can Machine Learning Advance Quantitative Asset Management?

Contact Info

Product

Resources

About