Modeling and Forecasting Persistent Financial Durations

Bibliographic record. Follow the original-source link for the publication.

Field	Value
Primary domain	Volatility
Other domains	—
Methods	Research Methods
Facets	—
Authors	Filip Zikes, Jozef Barunik, Nikhil Shenai
Published	2012-08-15
Source	arXiv Quantitative Finance History
Identifiers	arxiv:1208.3087
URL	Open original source

Editorial synthesis

Why it matters

The study brings a Markov-switching multifractal mechanism into financial durations and traces how duration persistence propagates to counts and realized volatility, linking event timing to volatility forecasting. (abstract:S1, abstract:S2, abstract:S3, abstract:S6)

Main author claims

• The authors introduce an MSMD duration model and claim that, despite exponential beta-mixing, it can generate highly persistent autocorrelation. (abstract:S1, abstract:S2)
• The authors study analytically and by simulation how duration persistence propagates to counts and realized volatility, and establish strong consistency and asymptotic normality for a Whittle-based QMLE. (abstract:S3, abstract:S4)
• The authors report: In an out-of-sample comparison using price durations for three major FX futures contracts, the authors report similar performance for MSMD and LMSD, both outperforming short-memory ACD models. (abstract:S5, abstract:S6, abstract:S7)

Data, method, or discussion scope

The material covers the model definition, mixing and estimator properties, analytical and simulated propagation, and an out-of-sample forecast comparison on three FX futures contracts. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7)

Main limitations

The empirical scope is three FX futures contracts, and the abstract omits the sample period, loss functions, and estimation uncertainty, so persistence of the advantage across markets and evaluation criteria is unknown. (abstract:S6, abstract:S7)

Relationships

• None recorded.