Continuous Futures Contracts Methodology for Backtesting

Abstract

This paper reviews the problem of futures contracts and backtesting. If the contracts are spliced together, the resulting backtest is wrong. Such a dataset would include artificial and non-existing jumps that would appear in the analysis as profits or losses. This paper aims to examine some possible solutions to the spliced futures problem. Firstly, it is essential to choose the date when the successive contracts are rolled and secondly, which adjustments would be made to the raw contract prices. Since there are many options for both key elements, it creates a broad set of possibilities. Moreover, there is no one best approach. Each algorithm has it is own pluses and minuses. We examine in the practice the first-of-month roll method and backward ratio adjustment method. Such an approach is simple and probably the best if we want to provide information about the percentual performance of the strategies.