Estimating correlation using intraday price data in financial markets

Auteurs

• Valentin Popov

Paramètres

• Manuels
• Manuels universitaires

En savoir plus sur le livre

This work focuses on correlation in financial markets, specifically the efficient estimation of correlation between asset pairs' returns using additional price information. This additional information includes daily highs and lows alongside open and close prices, as well as high-frequency data. Building on prior research, a new construct called Balanced Excess Return (BER) is introduced, representing the balance of the wicks in Japanese candlestick charts of open, high, low, and close (OHLC) data. In a Brownian Motion (BM) model context, the correlation of BERs is shown to be closely linked to process correlation, highlighting their significance in estimating the latter. The study develops efficient and robust correlation estimators that improve upon those from previous research, primarily because the earlier estimators assumed known process variances. The work discusses the implications of this assumption by examining correlation estimation in a normal model. It distinguishes between two scenarios: one with only daily OHLC prices and another with high-frequency data. Extensive simulations are conducted to evaluate the efficiency and robustness of the proposed estimators, particularly under conditions of jumps and excess kurtosis, as seen in various processes. The practical relevance of these estimators is demonstrated through an empirical study of their statistical properties and an application in portfolio selection.

Achat du livre

• Année de publication: 2013