Paper: arXiv 2403.00819
Authors: Markus Bibinger, Nikolaus Hautsch, Alexander Ristig
Abstract
We propose methods to infer jumps of a semi-martingale, which describes long-term price dynamics, based on discrete, noisy, high-frequency observations. Different to the classical model of additive, centered market microstructure noise, we consider one-sided microstructure noise for order prices in a limit order book. We develop methods to estimate, locate and test for jumps using local minima of best ask quotes. We provide a local jump test and show that we can consistently estimate jump sizes and jump times. One main contribution is a global test for jumps. We establish the asymptotic properties and optimality of this test. We derive the asymptotic distribution of a maximum statistic under the null hypothesis of no jumps based on extreme value theory. We prove consistency under the alternative hypothesis. The rate of convergence for local alternatives is determined and shown to be much faster than optimal rates for the standard market microstructure noise model. This allows the identification of smaller jumps. In the process, we establish uniform consistency for spot volatility estimation under one-sided noise. Online jump detection based on the new approach is shown to achieve a speed advantage compared to standard methods applied to mid quotes. A simulation study sheds light on the finite-sample implementation and properties of the new approach and draws a comparison to a popular method for market microstructure noise. We showcase how our new approach helps to improve jump detection in an empirical analysis of intra-daily limit order book data.
Complexity vs Empirical Score
- Math Complexity: 8.5/10
- Empirical Rigor: 6.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper employs advanced stochastic calculus and extreme value theory, with dense theoretical derivations, placing it in the high math quadrant. It includes simulation studies and empirical analysis of limit order book data, demonstrating implementation readiness, but the focus remains heavily on theoretical optimality and proofs rather than a fully packaged backtesting framework.
Research Flowchart
flowchart TD
A["Research Goal<br>Jump Detection in Limit Order Books"] --> B["Data Input<br>High-Frequency Order Prices"]
B --> C["Methodology<br>Local Minima of Best Ask Quotes"]
C --> D{"Key Computation<br>Estimate & Locate Jumps"}
D --> E["Local Jump Test<br>Consistent estimation of jump sizes/times"]
D --> F["Global Jump Test<br>Asymptotically optimal via EVT"]
E & F --> G["Outcomes<br>Faster detection of small jumps"]
G --> H["Validation<br>Simulation & Empirical Analysis"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes methods to infer jumps of a semi‑martingale from discrete, noisy, high‑frequency observations, including a global jump test with optimal asymptotic properties.
Source sentence
We propose methods to infer jumps of a semi‑martingale, which describes long‑term price dynamics, based on discrete, noisy, high‑frequency observations.
Assumes 4
- One‑sided, lower‑bounded microstructure noise with cdf Fη(x)=ηx+O(1) as x↓0. model
Source sentence
All (ϵi)0≤i≤n have a cumulative distribution function (cdf) Fη which satisfies Fη(x) = ηx 1 + O(1) , as x ↓0 .
- Noise is m_n‑dependent with Cov(ϵk,ϵl)=0 for |k−l|>m_n and m_n=O(log(n)). statistical
Source sentence
The noise is m_n‑dependent, such that Cov(ϵk, ϵl) = 0, for |k −l| > m_n, with a sequence m_n = O(log(n)).
- Observation times t_i^n converge weakly to a limit measure with a continuous, strictly positive Lebesgue density. data
Source sentence
... weakly, µS_n w→µS, to a limit measure µS which has a continuous, strictly positive Lebesgue density f_S on [0, 1].
- Noise distribution need not have finite moments. statistical
Source sentence
the developed asymptotic theory for the LOMN‑model does not require the existence of moments of the noise.
Data
- LOBSTER first‑level limit order book data order-book commercial
Source sentence
We use first‑level limit order book data of ask and bid quotes at the highest possible frequency from the LOBSTER database
Costs
Not applicable
Finds 4
- The convergence rate for local alternatives under one‑sided noise is n^{-1/3}, faster than the n^{-1/4} rate for standard MMN, allowing detection of smaller jumps. theoretical
Source sentence
The rate of convergence for local alternatives is determined and shown to be much faster than optimal rates for the standard market microstructure noise model.
- The global LOMN‑based jump test has higher power than comparable tests under MMN, especially for smaller jumps. performance
Source sentence
For fixed jump size and number of observations, the power of our test outperforms the power of tests under MMN.
- Empirically, more jumps are identified when using local minima of ask quotes and local maxima of bid quotes compared to mid‑quote methods. other
Source sentence
We find more jumps in the data study based on local minima of ask and local maxima of bid quotes.
- The pulverization effect of jumps caused by pre‑averaging in MMN is avoided using block‑wise local minima in the LOMN framework. theoretical
Source sentence
the effect of pulverization of jumps by pre‑averages reported in Mykland and Zhang (2016) under MMN
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 16 related papers). Coverage is partial, so this is neither replication nor refutation.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.
Cited References (57)
- Statistical inference for rough volatility: Minimax theory (2024) — link
- Optimal Nonparametric Range-Based Volatility Estimation (2024) — link
- A note on the Gumbel convergence for the Lee and Mykland jump tests (2023) — link
- High frequency market making: The role of speed (2023) — link
- Inference on the intraday spot volatility from high-frequency order prices with irregular microstructure noise (2023) — link
- Reading the Candlesticks: An OK Estimator for Volatility (2022) — link
- Gumbel convergence of the maximum of convoluted half-normally distributed random variables (2021) — link
- Volatility measurement with pockets of extreme return persistence (2021) — link
and 49 more
- High-frequency factor models and regressions (2020) — link
- Macroeconomic news, public communications, and foreign exchange jumps around U.S. and European financial crises (2020) — link
- News as sources of jumps in stock returns: Evidence from 21 million news articles for 9000 companies (2019) — link
- Jump factor models in large cross‐sections (2019) — link
- A ReMeDI for Microstructure Noise (2019) — link
- Large-dimensional factor modeling based on high-frequency observations (2019) — link
- Bootstrapping High-Frequency Jump Tests (2018) — link
- Estimating fast mean-reverting jumps in electricity market models (2018) — link
- Systemic Co-Jumps (2017) — link
- Mixed-scale jump regressions with bootstrap inference (2017) — link
- On high frequency estimation of the frictionless price: The use of observed liquidity variables (2017) — link
- Testing if the market microstructure noise is fully explained by the informational content of some variables from the limit order book (2017) — link
- Adaptive estimation of continuous-time regression models using high-frequency data (2017) — link
- Estimation of the discontinuous leverage effect: Evidence from the NASDAQ order book (2017) — link
- A Unified Approach to Volatility Estimation in the Presence of Both Rounding and Random Market Microstructure Noise (2017) — link
- Volatility, Information Feedback and Market Microstructure Noise: A Tale of Two Regimes (2017) — link
- The drift burst hypothesis (2016) — link
- Efficient Estimation of Integrated Volatility Incorporating Trading Information (2016) — link
- High-Frequency Financial Econometrics (2016) — link
- Between Data Cleaning and Inference: Pre-Averaging and Robust Estimators of the Efficient Price (2016) — link
- High Frequency Trading (2014) — link
- Estimating the Spot Covariation of Asset Prices—Statistical Theory and Empirical Evidence (2014) — link
- Volatility estimation under one-sided errors with applications to limit order books (2014) — link
- Inference on the Lévy measure in case of noisy observations (2014) — link
- Time endogeneity and an optimal weight function in pre-averaging covariance estimation (2014) — link
- Adaptive function estimation in nonparametric regression with one-sided errors (2013) — link
- Estimating the Quadratic Covariation Matrix from Noisy Observations: Local Method of Moments and Efficiency (2013) — link
- Testing for jumps in noisy high frequency data (2012) — link
- Jumps in equilibrium prices and market microstructure noise (2012) — link
- Discretization of Processes (2011) — link
- Intraday jumps and US macroeconomic news announcements (2011) — link
- Volatility Jumps (2011) — link
- A New Approach for the Dynamics of Ultra-High-Frequency Data: The Model with Uncertainty Zones (2011) — link
- Testing for Jumps in a Discretely Observed Process (2009) — link
- Jumps in Financial Markets: A New Nonparametric Test and Jump Dynamics (2008) — link
- Designing Realised Kernels to Measure the Ex-Post Variation of Equity Prices in the Presence of Noise (2008) — link
- Sampling Returns for Realized Variance Calculations: Tick Time or Transaction Time? (2006) — link
- Realized Variance and Market Microstructure Noise (2006) — link
- Econometrics of Testing for Jumps in Financial Economics Using Bipower Variation (2005) — link
- How Often to Sample a Continuous-Time Process in the Presence of Market Microstructure Noise (2005) — link
- Distribution Function Inequalities for Martingales (1973) — link
- Extreme Values in Samples from $m$-Dependent Stationary Stochastic Processes (1954) — link
- Change Point Detection in Beta Process with High Frequency Data (2023) — link
- Volatility coupling (2021) — link
- Change points and uniform confidence for spot volatility (2020)
- Time-varying continuous and jump betas: The role of firm characteristics and periods of stress (2017) — link
- Jump Regressions ∗ (2015) — link
- Roughing up Beta: Continuous vs. Discontinuous Betas, and the Cross-Section of Expected Stock Returns (2014) — link
- Extreme value theory : an introduction (2006) — link
0 of 57 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.