2021/01/15 by Murilo Sibrao Bernardini, Bernardini, Murilo Sibrao, Paulo André Lima de Castro +1 · 1 citation
Economics, Econometrics and Finance · Decision Sciences · #Financial Markets and Investment Strategies #Stock Market Forecasting Methods #Complex Systems and Time Series Analysis
paper · pdf · doi:10.48550/arxiv.2101.07217
In this paper, we propose a method for evaluating autonomous trading\nstrategies that provides realistic expectations, regarding the strategy's\nlong-term performance. This method addresses This method addresses many\npitfalls that currently fool even experienced software developers and\nresearchers, not to mention the customers that purchase these products. We\npresent the results of applying our method to several famous autonomous trading\nstrategies, which are used to manage a diverse selection of financial assets.\nThe results show that many of these published strategies are far from being\nreliable vehicles for financial investment. Our method exposes the difficulties\ninvolved in building a reliable, long-term strategy and provides a means to\ncompare potential strategies and select the most promising one by establishing\nminimal periods and requirements for the test executions. There are many\ndevelopers that create software to buy and sell financial assets autonomously\nand some of them present great performance when simulating with historical\nprice series (commonly called backtests). Nevertheless, when these strategies\nare used in real markets (or data not used in their training or evaluation),\nquite often they perform very poorly. The proposed method can be used to\nevaluate potential strategies. In this way, the method helps to tell if you\nreally have a great trading strategy or you are just fooling yourself.\n