highfrequency

Background

The highfrequency package is the go-to package for the analysis of intraday price data. The package was created as a merger of the packages RTAQ and realized in 2012. The package underwent a major rewrite last year improving the process and stability of the package but is in need of additional functionality that reflects current developments in financial econometrics and empirical finance.

Related work

There is the HighFreq package that provides some functionality to aggregate trades and quotes data but which is not available on CRAN.

Details of your coding project

New data management tools

• Euronext data support for facilitating research on European stock price determinants
• Aggregation is now based on calendar time. Business time aggregation is needed (number of ticks, volume increment, volatility increment)
• The refreshTime function is quite slow. This function will be optimized to run faster.

Improved functionality to analyze high frequency data:

• Estimators for the serial correlation in microstructure noise (Jacod, J., Li, Y., & Zheng, X. (2017). Statistical properties of microstructure noise. Econometrica, 85(4), 1133-1174. Li, Z. M., & Linton, O. B. (2019). A ReMeDI for microstructure noise.)
• Calculation and evaluation of jump matrix based on intraday tests (Li, J., Todorov, V, Tauchen, G. & Lin, H. 2019. Rank Tests at Jump Events. Journal of Business and Economic Statistics. 37: 312-321.)
• Cholesky-based realized covariance estimation (Boudt, K., Laurent, S., Lunde, A., Quaedvlieg, R., & Sauri, O. (2017). Positive semidefinite integrated covariance estimation, factorizations and asynchronicity. Journal of econometrics, 196(2), 347-367.)
• Jump tests that are robust to drift burst.
• Estimate lead/lag relationships. See https://github.com/philipperemy/lead-lag. Provides an implementation in Python/Cython, based on: https://arxiv.org/pdf/1303.4871.pdf
• High-frequency RV/BV/jump test (Christensen, K., Oomen, R. C., & Podolskij, M. (2014). Fact or friction: Jumps at ultra high frequency. Journal of Financial Economics, 114(3), 576-599.; Lee, S. S., & Mykland, P. A. (2012). Jumps in equilibrium prices and market microstructure noise. Journal of Econometrics, 168(2), 396-406.)
• The drift burst test (Christensen, K., Oomen, R. C., & Renò, R. (2018). The drift burst hypothesis. Working paper).
• Correction for spurious jumps (Bajgrowicz, P., Scaillet, O., & Treccani, A. (2016). Jumps in high-frequency data: Spurious detections, dynamics, and news. Management Science, 62(8), 2198-2217.)
• Decomposition of RCov in realized semicovariance (Bollerslev, T., Patton, A. J., Li, J., & Quaedvlieg, R. (2019). Realized semicovariances. Econometrica)

Functionality to simulate high frequency data:

• Simulate high-frequency price data under various configurations for the spot volatility and presence of jumps and microstructure noise. This could also include multivariate simulation with (possibly time-varying) correlations between the price paths.

One simulation method could be that of (Bollerslev, Patton, Quaedvlieg (2015). Exploiting the errors: A simple approach for improved volatility forecasting (Appendix))

Expected impact

The changes to the package reflect the requests of different users of the highfrequency package. Addressing those needs will thus be useful for the R community.

Mentors

Kris Boudt, Onno Kleen, Nabil Bouamara.