Modeling price clustering in high-frequency prices

DOI10.1080/14697688.2022.2050285zbMATH Open1498.91418arXiv2102.12112OpenAlexW3132635290MaRDI QIDQ5039627FDOQ5039627

Authors: Vladimír Holý, Petra Tomanová 

Publication date: 30 September 2022

Published in: Quantitative Finance (Search for Journal in Brave)

Abstract: The price clustering phenomenon manifesting itself as an increased occurrence of specific prices is widely observed and well-documented for various financial instruments and markets. In the literature, however, it is rarely incorporated into price models. We consider that there are several types of agents trading only in specific multiples of the tick size resulting in an increased occurrence of these multiples in prices. For example, stocks on the NYSE and NASDAQ exchanges are traded with precision to one cent but multiples of five cents and ten cents occur much more often in prices. To capture this behavior, we propose a discrete price model based on a mixture of double Poisson distributions with dynamic volatility and dynamic proportions of agent types. The model is estimated by the maximum likelihood method. In an empirical study of DJIA stocks, we find that higher instantaneous volatility leads to weaker price clustering at the ultra-high frequency. This is in sharp contrast with results at low frequencies which show that daily realized volatility has a positive impact on price clustering.

Full work available at URL: https://arxiv.org/abs/2102.12112

Recommendations

• Volatility clustering in financial markets: empirical facts and agent-based models
• High-Frequency Volatility and Liquidity
• Volatility clustering in agent based market models
• Modeling the coupled return-spread high frequency dynamics of large tick assets
• On the study of two models for integer-valued high-frequency data

zbMATH Keywords

high-frequency datageneralized autoregressive score modelprice clusteringdouble Poisson distribution

Mathematics Subject Classification ID

Applications of statistics to actuarial sciences and financial mathematics (62P05) Portfolio theory (91G10) Financial markets (91G15)

Cites Work

• Generalized autoregressive conditional heteroscedasticity
• A model for integer-valued time series with conditional overdispersion
• Double Exponential Families and Their Use in Generalized Linear Regression
• Autoregressive Conditional Duration: A New Model for Irregularly Spaced Transaction Data
• Designing Realized Kernels to Measure the ex post Variation of Equity Prices in the Presence of Noise
• The Econometrics of Ultra-high-frequency Data
• Realized kernels in practise : trades and quotes
• Dynamic models for volatility and heavy tails. With applications to financial and economic time series
• Information-theoretic optimality of observation-driven time series models for continuous responses
• Price clustering in bitcoin
• Integer-valued Lévy processes and low latency financial econometrics
• Extended Poisson INAR(1) processes with equidispersion, underdispersion and overdispersion
• Underdispersion models: models that are ``under the radar
• On the probability of default in a market with price clustering and jump risk

Cited In (4)

• Modeling discrete stock price changes using a mixture of Poisson distributions
• Time-varying arbitrage and dynamic price discovery
• Modeling microstructure price dynamics with symmetric Hawkes and diffusion model using ultra-high-frequency stock data
• Modeling the coupled return-spread high frequency dynamics of large tick assets

Uses Software

• BRENT

This page was built for publication: Modeling price clustering in high-frequency prices

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5039627)