Stochastic regression and its application to hedging in finance (Q1042957)

The authors consider stochastic regression between a stochastic process \(\Xi\), that is the sum of residual noise and stochastic integrals with respect to semimartingales, and these semimartingales. From the financial point of view, semimartingales are considered as price processes of the collection of securities, and \(\Xi\) as the price process of a newly added security. The motivation comes from statistical risk management in financial markets. The purpose of the paper is to investigate how to reduce the remaining unhedgeable risk, measured by the quadratic variation of the residual process. Some practical recommendations are given, how to choose hedging securities.