Using Primary and Secondary Market Movements to Construct an Optimal Time-Series Momentum Strategy : A Replication Study

Abstract

Time series momentum (TSM) strategies is a topic that has been analyzed in numerous academic journals; often the results of the studies imply that TSM outperforms the benchmark (buy-and-hold strategy). Nevertheless, most of the research covers primary trends as proposed by the Dow Theory. We implement a new TSM strategy that in addition to the primary trends, also considers the secondary trends in the Dow Theory. This TSM strategy is then applied to various look-back periods(speed), including predetermined static speeds, and dynamic speeds. The latter in which we use back-testing to find optimal speeds for different market states (bull, bear, correction, and rebound), and implement the speeds in subsequent periods with forward-testing. The TSM strategies are applied on international market indices, and the Sharpe ratio for each strategy reveals that the dynamic speed strategies dominate in terms of performance.