No matter how you look at it, the financial markets are a mathematical calculation to create specific value’s. That is why we have decided to draw a statistical picture on the current market attributes in the hope of giving our readers some important insights.
The way we will do this is by comparing the current price action with historical patterns, understanding these figures helps us get an edge in our day to day trading. But to do this, first we need to objectively describe the current conditions to make valid references. We need to precisely know where we are now and how we got here to make a meaningful comparison.
You might have heard analysts repeatedly calling markets as overbought over the past year or so. The main reason for this was that starting from September 2011 to September 2014, the market (S&P500) put on a spectacular show climbing up some 77% (on a close to close basis) in a space of 1065 days. As the market rallied we did not see any noticeable correction which was a major source of dispute among market analysts.
To put this into perspective let’s go back, all the way to December 1950 and study the history of major bull and bear swings. For the purpose of this exercise, we only looked at moves (up or down) which were equal or greater than 10% measured from the closing prices. We believe any move less than this threshold can be mixed with market noise and may not have true economic significance.
We found 21 major upswings plus 22 downswings in the study period. We then broke those into different sizes and durations. The results are interesting and are presented in the graphs below.
Graph 1, shows the distribution of upswing sizes once they get past a 10% threshold. According to the graph, there is an 80% chance that upswings reverse before market grows by 70%
Graph 2, shows the length of the period of upswings once they get past the 10% threshold. According to the graph, there is a 76% chance that an upswings fade out before they are a 1000 days old.
So based on the above, market behaviour in the past two years (77% increase in 1065 days ) was outside of the norm and excessive. This powerful rally was mainly driven by the US Fed which we have discussed in two of our previous articles.
As you know markets have now come off from their September highs and have nearly posted a 10% drop on a closing basis (this article was written when the S&P 500 closed at 1904). So our next step is to look back at the history again and study major down swings which have occurred after such rallies.
Graph 3, shows how the size of downswings are distributed. According to this graph, 64% of downswings end before reaching the 20% mark. Or we could say that there is a 64% chance that the current correction ends before turning into a crisis. Therefore, historically speaking we should not be far away from another intermediate low for now.
Graph 4, shows how long the downswings will take before they diminish. According to this graph, 27% percent of downswings of 10% or more will die before they get 100 days old. This means that the correction started last month, if it was to continue then it may still have some time left.
To help strengthen our analysis we also looked at the quarterly returns. Quarterly returns are closely followed by fund managers and can cause considerable flow of funds in and out of equity markets.
For the purpose of this analysis, we studied S&P quarterly returns (measured from closing prices) from December 1927 up to 21 October 2014.This gave us a total of 348 observations to work with.
In total, 215 quarters have recorded a price increase with the rest (133) showing price declines. Graph 5 shows the distribution of these returns. According to this graph, The market will move between -4% to 4% each quarter about 50% of the time.
Now given that the markets have recently corrected, we decided to dig deeper and only look at the quarters with negative returns .What we found was that under normal circumstances, around 47% of negative quarterly returns range between 0 to -4% . This means that the current market correction of -3.4% ( on a quarterly close to close basis) is not outside the norm and won’t be alarming in itself. This reinforces the message we found in graph 3.
Next, we examined market performance one and two quarters after it recorded a drop of less than 4%. We were curious to know what the chance of a rally is one and two quarters after a quarterly drop of less than 4% occurs. As it turns out , such quarterly drops do not cause subsequent rallies with great confidence.
Based on the table below, there is only 53% chance of rally in the next quarter. This is only a tad more than tossing a coin to predict the market. This indirectly reinforces the message we get under graph 4 which says the current correction can continue a little longer
|Chance of rally in the next quarter||53%|
|Chance of rally in the second quarter||54%|
To get another perspective, we looked at the valuations. We used the S&P 500 P/E ratio starting from 1954 as recorded by Bloomberg. For those who are not across with financial terms, P/E or Price to Earning ratio shows how inflated the prices are in relation to the underlying earnings. The higher the ratio the more risk of price bubble.
This ratio is different from Shiller P/E ratio which we discussed in one of our previous articles. Shiller makes adjustment for inflation and business cycles whereas this one doesn’t. Nevertheless this ratio is looked at by many professionals and that’s why we are including it in this article.
Based, on the graph below the most stable P/E ratio is between 16 -19 . Only 22% of the times market has recorded a reading of 22 or higher. Market is currently at a P/E of 17.18 which is not too high but not cheap as well.
Based on the above findings we can say that from a purely statistical point of view :
- The current correction is not alarming (at the moment) and has happened many times in the past.
- Historically speaking there is a 73% chance that this correction continues to at least December
- Quarterly returns are not dangerous and doesn’t seem to be generating excessive fear among fund managers
At Trade View we try to understand the mathematics behind the markets as everything in the markets relies on a calculation of numbers (earnings, growth, etc). The only real anomaly in the markets is the human factor.
If you would like to learn more about how Trade View builds Statistical models, please contact us and we will happy to discuss.
Most retail traders are now automating their trading models to build a systematic and disciplined approach in achieving their goals.
Build your very own automated trading system and receive a new model or concept straight from our Trading Desk every week. Get access to our Online Systems Building PLUS program now!