Objective research to aid investing decisions
Menu
Value Allocations for September 2019 (Final)
Cash TLT LQD SPY
Momentum Allocations for September 2019 (Final)
1st ETF 2nd ETF 3rd ETF

Trading Price Jumps

Posted in Commodity Futures, Currency Trading, Technical Trading

Is there an exploitable short-term momentum effect after asset price jumps? In his January 2017 paper entitled “Profitability of Trading in the Direction of Asset Price Jumps – Analysis of Multiple Assets and Frequencies”, Milan Ficura tests the profitability of trading based on continuation of jumps up or down in the price series of each of four currency exchange rates (EUR/USD, GBP/USD, USD/CHF and USD/JPY) and three futures (Light Crude Oil, E-Mini S&P 500 and VIX futures). For each series, he looks for jumps in prices measured at seven intervals (1-minute, 5-minute, 15-minute, 30-minute, 1-hour, 4-hour and 1-day). His statistical specification for jumps uses returns normalized by local historical volatility. He separately tests the last 4, 8, 16, 32, 64, 128 or 256 measurement intervals for the local volatility calculation, and he considers jump identification confidence levels of 90%, 95%, 99% or 99.9%. His trading system enters a trade in the direction of a price jump at the end of the interval in which the jump occurs and holds for a fixed number of intervals (1, 2, 4, 8 or 16). He thus considers a total of 6,860 strategy variations across asset price series. He divides each price series into halves, employing the first half to optimize number of volatility calculation measurement intervals, confidence level and number of holding intervals for each measurement frequency. He then tests the optimal parameters in the second half. He assumes trading frictions of one pip for currencies, and one tick plus broker commission for futures. He focuses on drawdown ratio (average annual profit divided by maximum drawdown) as the key performance metric. He excludes price gaps over weekends and for rolling futures contracts. Using currency exchange rate data during November 1999 through mid-June 2015, Light Crude Oil futures data during January 1987 through early December 2015, E-Mini S&P 500 futures during mid-September 1999 through early December 2015 and VIX futures during late March 2004 through early December 2015, he finds that:

  • For the trading strategy applied to currency exchange series with aggressive leverage:
    • A 1-minute measurement interval is consistently unprofitable due to trading frictions.
    • Best results are at 15-minute and 30-minute measurement intervals, with three of four series net profitable. USD/CHF produces the highest drawdown ratios (57% and 137%, respectively), followed by EUR/USD (43% and 33%) and GBP/USD (18% and 14%). However, USD/JPY produces loses (-3% and -0.2%).
    • A 1-hour measurement interval is generally unprofitable, except for a small net profit for USD/JPY.
    • For a 4-hour measurement interval, GBP/USD produces strong net profit, EUR/USD and USD/JPY are modestly net profitable and USD/CHF is unprofitable.
    • A 1-day measurement interval generates too few jumps for reliable inference.
  • For the trading strategy applied to futures series with aggressive leverage:
    • Light Crude Oil is unprofitable for 1-minute, 15-minute and 30-minute measurement intervals, and barely profitable for a 5-minute measurement interval. Drawdown ratios are attractive for 1-hour (21%) and 4-hour (22%) measurement intervals. There are not enough trades for a 1-day measurement interval for reliable inference.
    • E-Mini S&P 500 and VIX futures produce losses for all measurement intervals, suggesting a mean reversion (not momentum) strategy.
  • Less aggressive leverage mutes net profits for attractive scenarios.

In summary, evidence indicates that jump momentum trading may be attractive for some currencies for price measurement intervals of 15 to 30 minutes.

Cautions regarding findings include:

  • Assuming trade entry immediately at the end of a jump interval (no delay for calculations and trade execution) may not be realistic.
  • While in-sample versus out-of-sample testing avoids look-ahead bias, the study impounds data snooping (lucky measurement frequencies and lucky asset price series) by examining multiple strategy variations across multiple assets, thereby overstating expectations for the best-performing combinations.
  • It might be instructive to test whether in-sample optimization does any good (better than average out-of-sample results) by testing all variations with out-of-sample data.
  • Some traders may be uncomfortable with aggressive leverage.
  • Liquidity may diminish after price jumps, thereby increasing bid-ask spreads (trading frictions) and reducing trade capacities.
Why not subscribe to our premium content?
It costs less than a single trading commission. Learn more here.
Daily Email Updates
Login
Research Categories
Recent Research
Popular Posts