| Algorithmic trading, also known as program trading, the investment strategy is a programmable logic, then traded by means of computer investment decisions. In practice, an efficient trading system requires real-time analysis and processing of massive amounts of data, and then give trading signals. Under conditions in which the most important task is not to impose any a priori assumption that the probability of digging out the correlation between the various data streams and their evolution over time. This paper describes an efficient algorithm for a time-varying parameter estimation-Flexible least squares algorithm, the ordinary least squares algorithm is a variation, it can under no circumstances probability hypothesis quickly estimate the regression coefficients with time evolution. Secondly, in order to better understand the algorithm, this paper also discusses the FLS algorithm and the intrinsic link between the Kalman filter algorithm and give some improvements. Finally, we will FLS algorithm is applied to the high-frequency statistical arbitrage among, and on this basis to build a complete set of program trading systems, and have achieved good performance. |
PDF Full Text Request