The Monday Effect

Vladimir Daragan,  STTA Consulting Inc.

Table of Contents
1. Introduction
2. The Monday Effect
3. Day of the week (returns)
4. Day of the week (reversal points)
5. Conclusions

Introduction
Long time ago it was found that on Mondays the stock market continues its Friday's move. If on Friday one sees a market drop, then there is a high probability of further stock price drop on Monday. If Friday was good for stock prices, then one can expect continuation of the positive market move on Monday. This is so called the Monday effect in the stock market.

This effect is related to the weekend market analysis performed by millions of investors and traders. Many people think "linearly". They believe in trends. If they see rising stock prices, they expect continuation of this short-term trend. Their decision is simple: we should buy stocks on Monday to catch the train. If people see falling stock prices, they become pessimistic and decide to sell stocks on Monday.

During the last few years the stock market has changed. Millions of traders use on-line information services and discount brokers. Stock traders have a possibility to perform the stock market analysis every day and it takes minutes, not hours as it used to be ten years ago. How about the Monday effect? Is it still valid? What is the probability of this effect?

We have performed a statistical analysis of the US stock market for the period from 1989 to 2001 and have found many interesting phenomena. The Monday effect is the most popular and important. However, it would be interesting to check stock market returns for every day of the week. Is it worth selling or buying stocks on some particular day of the week? How about seasonal effects? What is the worst month of the year for tech stocks? For broader market? Are these data statistically reliable?

Let's start with the Monday effect.

The Monday Effect
Consider 12 year history (1989 - 2000) of NASDAQ-100 index. Let's describe how we calculate the probability of the Monday Effect. Let P(i) be an index value on some day i.

P(Thr)   - closing index value on some Thursday
P(Fri)   - closing index value on the next trading day (Friday)
P(Mon)   - closing index value on the next trading day (Monday)

If Friday (or Monday) is a Holiday, we did not count this day in our statistics.

The probability of the Monday effect we estimate as the ratio

Probability = N/ Ntotal

where Ntotal is the total number of considered events. N is the number of events when

[P(Fri) - P(Thr)]  * [P(Mon) - P(Fri)] > 0

i.e. price differences are both positive or negative.

Strictly speaking these ratios are not the probabilities as they are defined in the text books. However, these numbers can provide good estimation of these probabilities.

We have performed calculations of the probabilities of the Monday effect for every year and plotted distributions (histograms) of these numbers. Let us show the histogram for NASDAQ-100 index.

The Monday effect is very strong for this index. Only two years from 12 had the probabilities less than 50%. The next Table shows the values of the probabilities for every year.

One year (1997) had the effect probability greater than 70%!  The average value of this effect is equal to 61%.

We have performed the statistical analysis of this effect for many market indices. It was found that there are some group of stocks where this effect is very weak. The full analysis of the Monday effect we have published in our e-book How to Win the Stock Market Game.

Day of the week (returns)

Consider average returns of the market indices for every day of the week. We define returns on the day i as

Return = [P(i) - P(i-1)] / P(i-1) * 100%

Let us start with an analysis of the SP 500 and NASDAQ-100 indices. The next figure presents the average daily returns for every day of the week for the period from 1989 to 2000.

SP 500	NASDAQ-100

The error bars show the standard deviations of the returns. One can see the greatest returns fo SP 500 for this period were on Mondays. For NASDAQ-100 the largest return was on Wednesdays. The standard deviations are rather big and we can not make any conclusion about profitability of using special day of the week to buy or sell stocks.

The full analysis of the day of the week and the month of the year effects we have published in our e-book How to Win the Stock Market Game.

Day of the week (reversal points)

More interesting information can be obtained form the analysis of the market reversal points. We will calculate the probability of short trend continuation as we did for the calculations of probabilities of the Monday effect. If this probability for some day of the week is less than 50%, one can conclude that this day is the most probable day of reverse of the market short-term trends. Let us show these probabilities for NASDAQ-100 index.

One can see that on Tuesdays we have well defined minimum of the probabilities of the short-term market trend continuation. So, if there is a market drop on Friday and on Monday, one can expect the market reverse on Tuesday.

The full analysis of these effects for other market indices we have published in the e-book How to Win the Stock Market Game.

Conclusions

The Monday effect of continuation of short-term trends has been confirmed. This effect is strong for actively traded tech stocks. On Tuesdays one can expect the reverse of short-term market trends.

This is the results of the statistical analysis for the period from 1989 to 2001. We can not promise that these effects will occur in the future. The market is changing. However we hope that this information will help you to improve your trading performance.