---
title: Can Interval Data Improve Volatility Forecasts? Evidence from Foreign Exchange Markets
url: https://www.ml-quant.com/papers/ssrn/4785170/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: SSRN 4785170
source_url: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4785170
featured: 2024-04-10
citations: 0
topic: Derivatives & Volatility
---


# Can Interval Data Improve Volatility Forecasts? Evidence from Foreign Exchange Markets

The research explores the use of interval-valued data in foreign exchange markets to enhance volatility forecasts, utilizing threshold autoregressive interval models for four major exchange rates.

- Source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4785170
- Identifier: SSRN 4785170
- Released: 2024-04-05
- First featured: Quant Letter No. 44 (2024-04-10): https://www.ml-quant.com/issues/2024-04-10/
- Citations (Semantic Scholar): 0
- Published in: not yet
- Topic: Derivatives & Volatility

## Related

- [FX Forecast Volatility in Risk Management](https://www.ml-quant.com/papers/ssrn/5114727/): A dynamic Bayesian model using skewed distributions improves currency risk management and hedging strategies by better capturing financial data asymmetry.
- [TSMixer and Realized Volatility Prediction](https://www.ml-quant.com/papers/ssrn/4713756/): Stock Volatility Forecasting with Neural Networks: The TSMixer neural network model has proven to be more effective than traditional models in predicting stock market volatility, indicating a possible shift towards simpler models in the future.
- [SpotV2Net: Multivariate Intraday Spot Volatility Forecasting via Vol-of-Vol-Informed Graph Attention Networks](https://www.ml-quant.com/papers/arxiv/2401.06249/): Intraday Volatility Forecasting: The article introduces SpotV2Net, a new model for predicting intraday spot volatility using a Graph Attention Network, which has shown better accuracy in predicting Dow Jones Industrial Average index prices.
- [Harnessing Volatility Cascades with Ensemble Learning](https://www.ml-quant.com/papers/ssrn/4682793/): A modification to the base learner in bootstrap aggregation and boosting can significantly improve predictive accuracy in volatility forecasting, addressing substantial errors from parameter estimation.
- [CVA Hedging by Risk-Averse Stochastic-Horizon Reinforcement Learning](https://www.ml-quant.com/papers/ssrn/4673150/): The study uses risk-averse Reinforcement Learning for managing potential credit losses on a derivatives portfolio, proving its effectiveness through a numerical study for a portfolio consisting of a single FX forward contract.
- [TimesNet for Realized Volatility Prediction](https://www.ml-quant.com/papers/ssrn/4660025/): The study shows that the TimesNet model is effective in predicting stock volatility, particularly during extreme market movements, making it a strong neural network benchmark in volatility research.