Skip to content
All library documents

Futures Backtests with Negative Back-Adjusted Prices

Article Quant Q&A · Author: flexington

Summary

The document raises a practical question about trend-following backtests using ten-year bond futures. Panama-style back-adjustment can create negative historical prices, making ordinary simple or logarithmic percentage returns unusable for those observations. The author asks how to calculate portfolio profit and loss and size positions when the adjusted series crosses zero.

It does not provide an answer, method, worked example, or evidence. The useful takeaway is the modeling issue itself: back-adjusted prices are a constructed continuous series, and their levels need not behave like tradable contract prices. Any backtest approach must therefore define how to derive PnL and exposure from actual contract price changes and contract specifications, rather than relying on percentage returns from the adjusted level. The document leaves position sizing and portfolio construction unresolved, so it serves as a research question rather than an instructional treatment.

Key ideas

  • Panama back-adjustment can produce negative historical futures prices.
  • Percentage returns are not defined in the usual way when the price series crosses zero.
  • The author asks how to calculate PnL and position size for a futures trend strategy.
  • The document presents the problem but offers no proposed solution or supporting evidence.

Tags

Full text
# Futures Backtest


# Futures Backtest












Trying to get my head around a simple futures backtest, namely trend. What I am struggling with is after back-adjusting (Panama canal) 10 year bond futures, I'm left with negative prices in the series. How on earth can I use this in a portfolio context? I'm familiar with using returns (both simple and log) but that clearly can't apply in this case. How do I go about position sizing and constructing a PnL from these negative prices?

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.