Optimal Sampling for the Detection of Market Microstructure Noise
| Authors | |
|---|---|
| Year of publication | 2015 |
| Type | Article in Proceedings |
| Conference | European Financial Systems 2015. Proceedings of the 12th International Scientific Conference |
| MU Faculty or unit | |
| Citation | |
| Field | Management and administrative |
| Keywords | market microstructure noise; optimal sampling; LM test |
| Description | Volatility patterns and its dynamics are the core measures of risk in the financial theory. However, given the algorithmic nature of modern securities trading, frequently used parametric volatility models should be used with great caution when applied on high frequency data. Modelling volatility in high frequency data is fairly complex since such data contains a disruptive volatility component, which only occurs in this kind of data and is not observable in lower frequency data. This phenomenon is usually called market microstructure noise. It is mostly caused by bid ask bounce, so its presence is not so significant in assets with lower spreads. This paper focuses on the comparison of two approaches and simulations to identify market microstructure noise and derive optimal samples for measuring volatility. These tests are implemented on the high frequency trading data from the German Stock Exchange. Our paper provides high-frequency data optimal sampling solutions for risk managers and active investors. |
| Related projects: |