The Day of the Week Effect

Bishop’s University Department of Economics The “Day-of-the-Week” Effect: Analysis of Trends in the Daily Returns of Copper and Aluminum Lucas Zawislak and Jennifer Lee Dr. M. Vigneault Applied Economic Analysis March 15th, 2013 Introduction According to the neoclassical school of economics, asset markets are assumed to be both efficient and random.

These two assumptions are the base from which two neoclassical theories are derived: 1) “The Efficient Market Theory” infers that the market is remarkably adept in its utilization of information; while 2) “The Random Walk Model” infers that accurate predictions of outcome cannot be made on the basis of historical data. In summation, it is assumed that the price behavior of assets is essentially random, and all relevant information is almost immediately incorporated into price. There are two key elements, in reference to market participants or decision makers, engrained in the neoclassical position.

First, it is presumed that the decision maker is rational and therefore makes decisions using the expected utility function. Second, this position reasons that each decision maker has access to, and uses, full information about the fundamental valuations of assets. Consequently, the market should be comprised of distinctly independent, fully informed and rational decision makers. Contrary to the neoclassical belief’s studies have uncovered irregularities, in asset returns, over specific ranges in time, specifically over the days of the week.

This observed anomaly is commonly referred to as “The-Day-of-the-Week-Effect” which challenges the notion of market efficiency and randomness. It proposes that the distribution of returns may vary according to the day of the week. The most distinct characteristic of this anomaly is a pattern of positive returns on Friday coupled with negative returns on Monday, also known as “the weekend effect”. Purpose and Motivation The objectives of this study are to determine if there is evidence of the day-of-the week-effect in the weekly price fluctuations of both Copper and Aluminum.

More specifically, we will determine if the assets returns are dependent on the day of the week in which they are generated. If this is proven true, it will have implications on the behavior of market participants in regards to the trading of these commodities. It would be difficult to directly and consistently exploit this effect each week, due to high transaction costs. The situation in which this could be best exploited would be when there are plans to add one of these commodities to a portfolio, due to some strategic objective.

In this case it would be advantageous to be aware of the effect and know exactly which day of the week the prices would be at their lowest. As I mentioned above, this anomaly will be tested against two base metals (commodities): copper and aluminum. Copper is the third most widely used metal in the world, and is highly versatile. It is a base metal used in building construction, power generation, transmission, electronic product manufacturing, and the production of industrial machinery and transportation vehicles. Aluminum is a substitute for copper and is used in many of the same applications.

Though the two metals are similar in application aluminum is a much cheaper alternative. When you familiarize yourself with the uses of both metals it becomes evident that they are essential to urban modernization. The demand for base metals is primarily fueled by economic growth, and though economic growth in the western hemisphere has slowed, countries such as China and India are experiencing a significant upward trend. Base metals are vital to this growth. On account of this demand, copper is in decreasing supply and due to uncertainty about future supply; this is likely to translate into price volatility.

When making a purchase decision this volatility can be offset by the knowledge of the price trends. Aluminum is still in good supply and due to its likeness to copper its demand is increasing. Method We have collected data on Copper and Aluminum prices, as reported on the London Metal Exchange, from January 2nd 2009 to February 15th 2013. The standard OLS method will be used to test the day-of-the-week effect in each of the commodities returns by regressing the data of the returns on the five daily dummy variables.

The regression model below will be the base from which all analysis will take place. Essentially the commodity prices will be the dependent variables in the regression, while time will be the independent variable. Regression Model I: Ri=the daily yied of the asset D1=1 if Monday;=0 otherwise D2=1 if Tuesday;=0 otherwise D3=1 if Wednesday;=0 otherwise D4=1 if Thursday;=0 otherwise D5=1 if Friday;=0 otherwise **Null Hypothesis of Interest: Daily Return Equation Rt=(PtPt-1-1)*100 Descriptive Statistics The descriptive statistics reflect the fore mentioned metal profiles. On verage copper returns are 43% higher than that of Aluminum. In terms of standard deviation the returns for both are quite similar. Both graphs indicate increasing volatility of returns, yet this is much more prominent for copper. This pattern supports my previous statement indicating decreasing supply and increasing demand as a source of volatility. The large range given by the minimum and maximum returns is another indication of the volatility of returns for both metals Works Cited Berument, M. , and Nukhet Dogan. “Stock Market Return And Volatility: Day-Of-The-Week Effect. Journal Of Economics & Finance 36. 2 (2012): 282-302. Business Source Complete. Web. 12 Mar. 2013. Boudreaux, Denis, Spuma Rao, and Phillip Fuller. “An Investigation Of The Weekend Effect During Different Market Orientations. ” Journal Of Economics & Finance 34. 3 (2010): 257-268. Business Source Complete. Web. 12 Mar. 2013. Derbali, Abdelkader, and Noureddine Khadraoui. “Day Of The Week Effect On Assets Return: Case Of The Stock Exchange Of Casablanca. ” Journal Of Business Studies Quarterly 3. 1 (2011): 274-283. Business Source Complete.

Web. 15 Mar. 2013. Hassan Chowdhury, Shah Saeed, and Rashida Sharmin. “Does Cross-Sectional Risk Explain Day-Of-The-Week Effects In Bangladesh Stock Market?. ” International Research Journal Of Finance & Economics 93 (2012): 84-94. Business Source Complete. Web. 15 Mar. 2013. Ulussever, Talat, Ibrahim Guran Yumusak, and Muhsin Kar. “The Day-Of-The-Week Effect In The Saudi Stock Exchange: A Non-Linear Garch Analysis. ” Journal Of Economic & Social Studies (JECOSS) 1. 1 (2011): 9-23. Business Source Complete. Web. 15 Mar. 2013.