Forecasting the volatility of theAustralian dollar using high‐frequency data:Does estimator accuracy improve forecast evaluation?

Abstract

We compare forecasts of the volatility of the Australian dollar exchange rate to alternative measures of ex post volatility. We develop and apply a simple test for the improvement in the ability of loss functions to distinguish between forecasts when the quality of a volatility estimator is increased. We find that both realized variance and the daily high–low range provide a significant improvement in loss function convergence relative to squared returns. We find that a model of stochastic volatility provides the best forecasts for models that use daily data, and the GARCH(1,1) model provides the best forecast using high‐frequency data.

Mixed-Stable Models: An Application to High-Frequency Financial Data

“…In the present research, we have examined the log-likelihood target function for real financial data. Target function (6) was calculated for returns series Allianz SE of size 7385, which was obtained using the previous-tick interpolation (1) with the time step ∆t = 1000. The ML solution vector for this data set is α = 1.541470, β = 0.004055, µ = 0.000005, σ = 0.001241.…”

“…The main applications belong to financial engineering, ranging from risk management to options hedging, transaction execution, portfolio optimization, and forecasting. Thus, Bailey and Steeley [1] compared forecasts of the volatility of the Australian dollar exchange rate to alternative measures of ex post volatility, using high-frequency data. Degiannakis and Filis [2] examined the importance of combining high-frequency financial information, along with the oil market fundamentals, to gain incremental forecasting accuracy for oil prices, showing that although the oil market fundamentals are helpful for long-run forecasting horizons, the combination of the latter with high-frequency financial data significantly improve oil price forecasts.…”

Volatility forecasting revisited using Markov‐switching with time‐varying probability transition

“…In our study, we employ the superior volatility measures, realized volatility (RV), which has been recorded in the numerous existing literature (Bailey & Steeley, 2019;Luo & Ji, 2018;Ma et al, 2017;Ma, Lu, et al, 2019;Ma, Zhang, Wahab, & Lai, 2019;Wang et al, 2012;Y. Wang, Ma, Wei, & Wu, 2016;X.…”