Our research will explore the intriguing phenomenon of the Pre-Holiday effect in commodities, particularly crude oil and gasoline. Historical data reveals a short-term price drift prior to major U.S. holidays, suggesting a trend in these markets. We hypothesize that this anomaly may be driven by increased demand for oil and its derivatives, such as gasoline, as people prepare for travel, often by car, during the holiday season. This seasonal behavior offers unique opportunities for market participants.
Contrary to the efficient market hypothesis (EMH), market anomalies consistently show statistical significance, challenging traditional views of economic theories. We are fond of the cusp of quantitative finance, which finds mathematical, statistical, and practical measures and invariably proposes ways to exploit them for financial benefit. One such anomaly is the Pre-Holiday Effect, which suggests that stock market returns on the trading days preceding holidays are significantly higher than on other days.
This phenomenon has been well-documented in equity and debt markets. We are now exploring it in commodity markets since we felt it is not documented well to its merit. Seasonality plays a significant role in commodities, influencing their prices and returns in predictable patterns due to exceptionally high tedious focus on supply and demand by producers and consumers in “hard (physically material delivered) assets,” be it agricultural or in natural resources domains.
However, our analysis has yet to see if the prevailing effect from the stock market translates into commodity and energy markets driven mainly by professional traders. So our question stands—will there be some pre-positioning from institutions for assuming increased fuel consumption during US public holidays?
The crude oil market is among the most critical and volatile markets globally. Numerous factors influence the price of oil, including
We hypothesize that:
The increased demand for crude oil before the U.S. holidays causes a short-term price drift.
Hypothesis #1
This phenomenon could be driven by higher holiday fuel consumption, leading to increased demand and, subsequently, higher prices. This is one explanation that may explain the fundamental underworkings of this anomaly.
We will use historical data from Yahoo Finance, with adjusted close taken into account. This data contains price reflections of splits, dividends, and/or capital gain distributions over time.
While crude oil can be traded in many variants ranging from CFDs to futures, we select the most common and well-liquid ETF variant to cover the effect. This way, it can also be replicated by individual investors and small-institutional ones to get a good idea of what can be expected in terms of performance stats and essential risk metrics. We have an educated guess that in lack of transaction costs, the strategy would perform similarly with front continuous contract Light Sweet Crude Oil Futures CL1 traded at NYMEX. However, margin requirements and direct market availability for exchange access for retail and non-professionals are limited, so we omit this option from the direct analysis.
So, we focus on two assets, U.S. ETFs:
The analyzed period ranges from their inception dates
to almost the present (2024-08-30).
As for our holiday definition, we focus on major U.S. holidays (when NYSE, NASDAQ, and AMEX are closed for the whole day; thus, no trading on them is taking place), such as:
As mentioned, we want to focus on price action on trading days immediately preceding and after selected (single) Holiday days. For initial considerations, we distinguish between D-5 and D+5, where D-1 is, for example, the last trading day before the holiday.
From the simple histogram showing distributions of returns per day, we try to understand what sequence to be long (buy) asset will yield the most return. In the following chart, we will display a bar chart of price movements from D-5 to D+5 (five days before and after the holiday):
We can see that the D-4 to D-1 period is the most profitable for holding a long position in the market. Thus, it would be optimal to establish a position four days before the Holiday and sell at the close of the day, just before the market closes.
We shall also calculate the t-statistics of the average performance during these days and depict them in a table:
In the following figure, the D-5 to D5 Equity Curves, if you were long only during one particular day, are shown:
This leaves us as a good precursor for forming a trading strategy that is to come next.
The trading rules are simple:
Weighting: Position sizing is based only on one instrument of the choice, as we will see later (USO or UGA).
Now, we show the results in full from our backtest in the form of
We see that the equity curve is steadily rising with acceptable volatility. Allocation of capital in only a few days a year is an interesting option when we incorporate such strategy into a well-diversified trading and/or investing portfolio.
Strategy experienced the most significant drawdown in the months leading to the 2022 Russian invasion of Ukraine, but surged afterwards.
How this strategy would perform on the UGA ETF was examined as an additional contribution to the analysis. And we go through the same steps as for USO before to report our results:
The pre-holiday effect is even more prevalent in the UGA ETF. Performance almost doubled, while risk metrics improved slightly. One concern for practitioners is that this ETF is a little bit less liquid with probable bigger bid-ask spreads, but it should not be a significant portion of the decrease in returns in terms of overall performance.
Interestingly, both ETFs experienced a selloff or profit-taking from long positions in D+1 days, but we did not investigate that further.
In conclusion, there is a summary and overview with takeaways from our findings and suggestions for trading crude oil around holidays:
Author: Cyril Dujava, Quant Analyst, Quantpedia
Are you looking for more strategies to read about? Sign up for our newsletter or visit our Blog or Screener.
Do you want to learn more about Quantpedia Premium service? Check how Quantpedia works, our mission and Premium pricing offer.
Do you want to learn more about Quantpedia Pro service? Check its description, watch videos, review reporting capabilities and visit our pricing offer.
Are you looking for historical data or backtesting platforms? Check our list of Algo Trading Discounts.
Or follow us on:
Facebook Group, Facebook Page, Twitter, Linkedin, Medium or Youtube
Share onLinkedInTwitterFacebookRefer to a friendBe first to know, when we publish new content
