ML Anthology

Universal Algorithm for Trading in Stock Market Based on the Method of Calibration

Vladimir V. V'yugin
ALT 2013 pp. 53-67
doi:10.1007/978-3-642-40935-6_5 /alt/2013/vaposyugin2013alt-universal/

Abstract

We present a universal method for algorithmic trading in Stock Market which performs asymptotically at least as well as any stationary trading strategy that computes the investment at each step using a continuous function of the side information. In the process of the game, a trader makes decisions using predictions computed by a randomized well-calibrated algorithm. We use Dawid’s notion of calibration with more general checking rules and some modification of Kakade and Foster’s randomized rounding algorithm for computing the well-calibrated forecasts. The method of randomized calibration is combined with Vovk’s method of defensive forecasting in RKHS. Unlike in statistical theory, no stochastic assumptions are made about the stock prices.

PDF ALT Semantic Scholar

Cite

Text

V'yugin. "Universal Algorithm for Trading in Stock Market Based on the Method of Calibration." International Conference on Algorithmic Learning Theory, 2013. doi:10.1007/978-3-642-40935-6_5

Markdown

[V'yugin. "Universal Algorithm for Trading in Stock Market Based on the Method of Calibration." International Conference on Algorithmic Learning Theory, 2013.](https://mlanthology.org/alt/2013/vaposyugin2013alt-universal/) doi:10.1007/978-3-642-40935-6_5

BibTeX

@inproceedings{vaposyugin2013alt-universal,
  title     = {{Universal Algorithm for Trading in Stock Market Based on the Method of Calibration}},
  author    = {V'yugin, Vladimir V.},
  booktitle = {International Conference on Algorithmic Learning Theory},
  year      = {2013},
  pages     = {53-67},
  doi       = {10.1007/978-3-642-40935-6_5},
  url       = {https://mlanthology.org/alt/2013/vaposyugin2013alt-universal/}
}