2012/05/16 by Vladimir V. V’yugin, V'yugin, Vladimir, В. Г. Трунов +1
Decision Sciences · Economics, Econometrics and Finance · #Advanced Bandit Algorithms Research #Complex Systems and Time Series Analysis #FOS: Computer and information sciences #FOS: Economics and business #Forecasting Techniques and Applications #Machine Learning (cs.LG) #Portfolio Management (q-fin.PM) #Stock Market Forecasting Methods
paper · pdf · doi:10.48550/arxiv.1205.3767
openalex publication_date 2012/05/16 · openalex created_date 2021/02/01 · openalex updated_date 2026/07/28
We present a universal algorithm for online trading in Stock Market which\nperforms asymptotically at least as good as any stationary trading strategy\nthat computes the investment at each step using a fixed function of the side\ninformation that belongs to a given RKHS (Reproducing Kernel Hilbert Space).\nUsing a universal kernel, we extend this result for any continuous stationary\nstrategy. In this learning process, a trader rationally chooses his gambles\nusing predictions made by a randomized well-calibrated algorithm. Our strategy\nis based on Dawid's notion of calibration with more general checking rules and\non some modification of Kakade and Foster's randomized rounding algorithm for\ncomputing the well-calibrated forecasts. We combine the method of randomized\ncalibration with Vovk's method of defensive forecasting in RKHS. Unlike the\nstatistical theory, no stochastic assumptions are made about the stock prices.\nOur empirical results on historical markets provide strong evidence that this\ntype of technical trading can "beat the market" if transaction costs are\nignored.\n