Practical academic paper related to #100 - Trading WTI/BRENT Spread Monday, 6 July, 2015

#100 - Trading WTI/BRENT Spread

Authors: Donninger

Title: The Poverty of Academic Finance Research: Spread Trading Strategies in the Crude Oil Futures Market


Harvey, Liu and Zhu argue that probably most of the Cross-Section of Returns literature is garbage. One can always try an additional factor and will find a significant Cross-Sectional result with enough trial and error. Lopez de Prado argues in a series of articles in a similar vein. Theoretically scientific results are falsifiable. Practically previous results and publications are checked only in rare occasions. Growth in a Time of Depth by Reinhart-Rogoff was the most influential economic paper in recent years. It was published in a top journal. Although the paper contained even trivial Excel-Bugs it took 3 years till the wrong results and the poor methodology was fully revealed. The reviewers did not check the simple spreadsheets. This paper analyzes a less prominent example about spread trading in the crude oil futures market by Thorben Lubnau. The author reports for his very simple strategy a long term Sharpe-Ratios above 3. It is shown that – like for Reinhart-Rogoff – one needs no sophisticated test statistics to falsify the results. The explanation is much simpler: The author has no clue of trading. He used the wrong data.

Notable quotations from the academic research paper:


"Thorsten Lubnau devotes in [7] an own page to the data question. But it is just a summary of well known facts like “The financial crisis starting in 2008 led to a sharp decline of oil prices”. The essential question of rolling is not addressed at all. The used time series is shown in figure 1 of the paper (see screen shot after Graphic-5). The figure matches the picture in Graphic-5. They are obviously unadjusted. One can consider the effect of rolling during the course of trading simulation. But everyone with a minimum of quantitative finance experience avoids this nasty and error prone task and uses backadjusted data. Packages like Unfair Advantage from CSI provide this feature automatically. Lubnau does not mention this point in his further considerations. He has ignored the effect of rolling at all.

All indexes start with 1,000,000$. The fixed-date rolling time series (red in Graphic-14) ends with a final value of 821,750$. The max. volume rolling time series (yellow) does in between considerable better, but the final result is with 781,020$ even worse. The different behavior in between is due to good/bad luck. There are large up- and downs. The swings within the Bollinger band are just noise. If the noise hits the band on the right side, the performance jumps up, if it is on the wrong side, the strategy nosedives. The green unadjusted time-series does much better. The final value is 3,019,900$. The overall win is +201,9%. The Sharpe-Ratio is 0.61 and the max. relative drawdown is 42.9%.This is not bad, but it is far from the > 3.0 Sharpe ratio claimed by Th. Lubnau.

Are you looking for more strategies to read about? Check

Do you want to see performance of trading systems we described? Check

Do you want to know more about us? Check