ALSO READ THIS : The McClellan Oscillator and Summation Index : JSE User Guide

The McClellan Oscillator and its companion Summation Index are among the most used breadth indicators in the financial services industry. PowerStocks Research was the first company in South Africa to offer these indicators going back 17 years to South African investors. During the many years we have computed, published and analysed these indicators for our own investment strategy and risk reduction, and those of our clients, we have come to learn additional, unpublished characteristics in their use that are extremely powerful.

Before we go onto those, a bit of background is required.

1.The Net Advance Line (NAL) is the daily difference between the number of advancing shares and declining shares on the JSE, excluding ETF's, derivatives and debt instruments. NAL is divided by the total number of issues to adjust for changes to the number of shares listed on the JSE over the decades. A daily NAL of 10 means advancing shares exceeded declining shares by 10% of the total issues traded.

2. The McClellan Oscillator (MCOS) is the difference between a 10% smoothed NAL (roughly equivalent to a 19-day exponential moving average) and a 5% smoothed NAL (very roughly equivalent to a 39 day EMA )

3. The McClellan Summation Index (MCSI) is the prior days' MCSI plus today's MCOS

We do not wish to discuss the relative merits of NAL, MCOS and MCSI here as these are covered is detail in our McClellan Breadth Indicators User Guide (the most comprehensive guide in the world on the use of these indicators). What we do wish to cover is additional merits in the use of these Indicators since we published our user guide 5 years ago.

Findings
Here are our additional findings based on research and our own practical use of the indicators:

1. It has traditionally been viewed as negative when the MCOS is below zero. When the MCOS is negative, the MCSI is by definition falling or trending down. However if the MCOS is negative and both, or even one of the 5% or 10% indexes are still above zero then these are FAR LESS negative connotations. Conversely, it is viewed as bullish when MCOS is above zero however it is far LESS BULLISH if either or both of the 5% or 10% indices are below zero.

2. It has traditionally been viewed as bearish for the stock market when the MCSI is below zero. However, it is far less bearish if the MCOS is positive at the same time (i.e. the MCSI is rising). Conversely it is viewed as bullish if the MCSI is above zero, but it is far less bullish if the MCOS is below zero (ie the MCSI is trending down.)

3. It is traditionally viewed as bearish is the MCSI is below zero, however it is far less bearish if the MCSI is trading above its medium-term simple moving average (the period of this average is proprietary to us.). Conversely it is considered bullish if the MCSI is above zero, however it is far less bullish if the MCSI is trending below its medium-term moving average.

4. A timing strategy derived from entering the JSE when the MCSI is above its medium-term moving average and exiting into non-interest bearing cash when it is below its medium-term moving average (a scheme originally proposed by Martin Pring in the 1970's) delivered 594% return in the 16 years since 1997 by being vested in the market only 51% of the time. This compares to the buy and hold's return of 440%. So the method delivered 1.35 times the performance of the buy+hold with HALF the risk (exposure). If one would include interest accrued for the periods the timing strategy was in cash, the return would balloon to over 740%

The McClellan Diffusion Index (MDI)
The MDI is an invention of PowerStocks Research that attempts to capture the  bullish/bearish connotations of the various components of a McClellan indicator set, in order to offer a "risk reading" on the current status of the indicators. It consists of seven calculations computed daily as follows:

1. MDI starts at zero
2. If MCOS > 0 then add 1 to MDI
3. If 5% index is > 0 then add 1 to MDI
4. If 10% index is > 0 then add 1 to MDI
5. If MCSI is > 0 then add 1 to MDI
6. If MCSI > medium term moving average then add 1 to MDI
7. Multiply final MDI result by 20

The MDI thus ranges from 0 (bearish) to 100 (maximum bullish) in steps of 20. The idea is that it suggests the percentage exposure you should have to the market. When it reads 0% you should be all-cash. When it reads 100% you should be all in the market and when it reads 60% you should be 40% cash and 60% in the JSE. The JSE All-share index together with the MCDI is plotted out below:


You can see when a major market top/bear market approaches the MCDI starts scaling back your exposure to the stock market. It is thus a method for you to adjust you equity allocation to the risk inherent in the markets.

A zoomed-in view of the MCDI covering the bull market that commenced in March 2009 is displayed below as at 17 Jul 2013. It is showing a 60% bullish signal.

Here is the view on the Great Bull Market leading up to the Great Recession of 2008. Note how we were kept on the right side of the bull market and had limited exposure to the Great Crashes of 2008. We didn't completely avoid all the damage, but we avoided the worst of it. The periods we were invested during the crash averaged only 40% but a big chunk of it we avoided by being in cash (Diffusion = 0). An equity curve of such an investment regime would have had far less draw downs and volatility associated with it.


Here is a view just before the above chart, showing how we skirted most of the carnage of the 2202/2003 crash and recession:


And here is the view for the period before the above chart, again demonstrating the ability of market breadth information to steer us clear of the worst of debilitating, vicious stock market crashes:


Now of course we have robust Econometric Models to assist us steer clear of recessions and bear markets, namely the SuperModel, the Long-Bond model, the BigMAC and the Cash/Equity Model (all of which are captured together in the Quattro Investment Model), but it is nice to be able to refer to the McClellan Diffusion Index together with these other models we have traditionally used. The MCDI offers a risk measurement purely from a market breadth point of view whereas the Econometric models look at all sorts of other things, so if anything the MCDI brings valuable information to the party for managing stock market risk.

ALSO READ THIS : The McClellan Oscillator and Summation Index : JSE User Guide
 
Make a Free Website with Yola.