Do mixed-data sampling models help forecast liquidity and volatility?

Помогают ли модели выборки смешанных данных прогнозировать ликвидность и волатильность?
Barbara Będowska-Sójka, Agata Kliber
2022-10-31

Granger causalityliquidity forecastingmixed-data sampling (MIDAS)vector autoregressive modelsvolatility forecasting
This paper aims to contribute to the existing studies on the Granger-causal relationship between volatility and liquidity in the stock market. We examine whether liquidity improves volatility forecasts and whether volatility allows the improvement of liquidity forecasts. The forecasts based on the mixed-data sampling models, MIDAS, are compared to those obtained from models based on daily data. Our results show that volatility and liquidity forecasts from MIDAS models outperform naive forecasts. On the other hand, the application of mixed-data sampling models does not significantly improve the performance of the forecasts of either liquidity or volatility based on a univariate autoregressive model or a vectorautoregressive one. We found that in terms of the forecasting ability, the VAR models and the AR models seem to perform equally well, as the differences in forecasting errors generated by these two types of models are not statistically significant.
1
MIDAS models produce liquidity and volatility forecasts that outperform naive forecasts.
2
Mixed-data sampling does not significantly improve liquidity or volatility forecasts relative to daily-data univariate autoregressive or vector autoregressive models.
3
The study examines bidirectional Granger-causal links between stock-market liquidity and volatility through their forecasting performance.
4
VAR and AR models perform similarly, with differences in forecasting errors not statistically significant.

Stock-market liquidity and volatility

The Granger-causal relationship between liquidity and volatility and the comparative forecasting performance of MIDAS, autoregressive, and vector-autoregressive models

Publication Details
Publication Date
2022-10-31
Journal
Publisher
ISSN
Cited by
0
Access Type
Author Information
Authors
Barbara Będowska-Sójka
Agata Kliber
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%