Please rotate your device to landscape mode to view the charts.

Time series reversal in trend-following strategies

Journal: European Financial Management

Date: 2023

Author: Jiadong Liu, Fotis Papailias

Abstract:
This paper empirically studies the reversal pattern following the formation of trend-following signals in the time series context. This reversal pattern is statistically significant and usually occurs between 12 and 24 months after the formation of trend-following signals. Employing a universe of 55 liquid futures, we find that instruments with sell signals in the trend-following portfolio ('losers') contribute to this type of reversal, even if their profits are not realised. The instruments with buy signals in the trend-following portfolio ('winners') contribute much less. A double-sorted investment strategy based on both return continuation and reversal yields to portfolio gains which are significantly higher than that of the corresponding trend-following strategy.

Link: Google Scholar

Background and Context

Research Focus

Study examines reversal patterns that occur after trend-following trading signals in financial markets using data from 55 liquid futures contracts from 1985-2015.

Methodology

Analyzes two types of trend-following strategies - Time Series Momentum (TSM) and Return Signal Momentum (RSM) - to identify and measure reversal effects.

Data Coverage

Dataset includes 24 commodity futures, 9 foreign exchange futures, 9 equity indexes, and 13 government bonds across major developed markets.

Time Series Reversal Occurs 12-24 Months After Signal Formation

  • Initial momentum profits in months 1-12 are followed by significant reversal in months 13-24
  • Reversal effect is strongest during the 13-24 month period
  • Pattern stabilizes after 24 months with no significant reversals

Contrarian Loser Portfolio Shows Strongest Reversal Effects

  • Contrarian loser portfolio generates highest returns of 8.4% annually
  • Past losers that later generate positive returns show strongest reversal
  • Winner portfolios show weaker or insignificant reversal patterns

Strategy Performance Declines with Longer Holding Periods

  • Returns gradually decrease as holding period increases from 1 to 24 months
  • One-month holding period strategies perform best
  • Benefits of trend-following are offset by subsequent reversals over longer periods

Trend-Following Reversal Strategies Outperform Traditional Approaches

  • $1 invested in contrarian loser strategy grows to $209.07
  • Realised winner strategy turns $1 into $143.82
  • Both significantly outperform traditional TSM approach ($16.63)

Portfolio Composition Remains Well-Diversified

  • Portfolio maintains balanced exposure across all four strategy components
  • Realised winner comprises slightly larger portion at 28%
  • Even distribution helps maintain diversification benefits

Contribution and Implications

  • Identifies specific 12-24 month window when time series reversal occurs, which is shorter than traditional cross-sectional reversals
  • Demonstrates that trend-following and reversal effects are separate phenomena, challenging existing behavioral theories
  • Develops new trading strategy combining trend-following and reversal effects that significantly outperforms traditional approaches

Data Sources

  • Reversal timing chart based on Figure 2 showing t-statistics of predictability tests
  • Portfolio performance data from Table 2 showing post-holding period returns
  • Holding period analysis based on Figure 3 return decay patterns
  • Strategy comparison using cumulative returns data from Figure 6
  • Portfolio composition visualized using proportions data from Figure 7