Treat outliers for risk models with due care!
Risk managers, pension funds, asset managers and banks nowadays use advanced models to assess the risk of investment portfolios. Much scientific progress has been made over the past decade to develop new techniques to measure the risk of such portfolios. In recent years scientists and professionals have started to use so-called high-frequency data to measure the risk. High-frequency data are frequent measurements of, for instance, stock prices or exchange rates. You can think of measurements every minute, every second, or in some cases even every millisecond. Such measurements often result in more accurate risk assessments than traditional daily measurements.
An important issue, however, is how to deal with so-called outliers in high-frequency data. An outlier is an anomalous measurement in the data. Think of a temporary crash in markets due to a faulty algorithm, a typo by a trader, or any other reason. Second, high-frequency based risk measures display `long memory’ behavior: the computed risk measure of 50 days ago is still related to the risk measure of today. Most of the current risk models do not take this property into account, while it enables the risk manager to forecast risk many days ahead.
In Opschoor, et al. (2018) we have developed a new technique to deal with such anomalous observations in high-frequency data. The core novelty of our approach is that anomalous events do not automatically inflate risk forecasts as in traditional models. Instead, the model trades off whether the increased risk is due to a true increase in risk, or to an incidental, anomalous event. Opschoor and Lucas (2018) extend this model by incorporating the ‘long memory’ property into the model.
The results show that with these new techniques risk forecasts are significantly better than with the most recent competing methods. Moreover, our method is relatively straightforward to implement.
Anne Opschoor is assistant professor at the department of Finance.
- Anne Opschoor, Pawel Janus, André Lucas, and Dick van Dijk (2018). New HEAVY Models for Fat-Tailed Covariances and Returns. Journal of Business and Economic Statistics (forthcoming)
- Anne Opschoor, and André Lucas (2018). Fractional Integration and Fat Tails for Realized Covariance Kernels (Tinbergen Institute Discussion Paper 16-069/IV)