Engineering Dimensionally Alternative Business Analysis
The Mind Set
This is the next step; taking hypothetical theory and showing what happens in practice! End of Day (EOD) Data is readily available and pretty useless in its isolated form; but when it is technically analysed in perspective with older data, and then the mind set of Technical Analysis comes to life!
Here is a visual representation of a security using EOD data. The blue spiles in the bottom section of the display are representative of turnover volume, and the upper section is the prices shown as 'candlesticks' - red when the price falls, and green if the price rises over the day.
Remember: Technical Analysis is not predictive - it is reactive. Insider information is predictive!
The secret of good Technical Analysis is to adjust the parameters of the indicators you are using so that they are in synchronism with the security being analysed. Most securities have a longer term trend with cyclic pattern of repetitive lows and highs - spaced by anything from less than days to several months - depending on the chip type and market forces.
It is far easier to align technical analysis indicators with long-term trends and have reasonably good accuracy than try to align indicators with short-term fluctuations.
General Positive Trend
The first stage is to filter out all those securities that are not on a medium/long term positive trend. This can be done by calculating a moving average of the recent End Of Day (EOD) market close prices that has a time constant of about half a year (109 trading days in this case) and compare that to the current close price. If the close price is greater than the moving average than the security has a high probability of being in a positive trend.
The above display is the same security as before (at another point in time) with the top section having a 109 day exponential moving average (EMA109) on it and the lower section of the display showing bars when the EMA109 is greater than the Close Price. If the EOD price is near in value to the EMA109 then the result can be indeterminate.
By including another (shorter/faster) moving average for example EMA21 as shown in above in the upper section, and comparing this with the EMA109 the decision has far less fluctuations, as shown in the lower section. By taking the difference between these two moving averages, this is called the Moving Average Convergence Divergence (MACD) Indicator, and this can be graphed as shown above in the middle section of the display.
Share prices are in fact a ratio of each other, and this needs to be visually represented using a semi-log price display as shown below. Also the MACD can be made very sensitive to fluctuations by adding a trigger (a moving average of the MACD), and these crossovers can be used for trading
The Candlestick graph now has a semi-log price axis and the prices are proportional as ratios and not arithmetically visualised. The Trigger line under the MACD essentially keeps track of the positive trend, and the decision graph on the lowest section essentially tracks the positive trends.
So the securities that we like are those that have a general positive trend - and Technical Analysis can really easily filter out all those that do not comply!
The second main criterion is that the securities being considered must be being actively traded, and again Technical Analysis cuts through this like a hit knife in butter. The simple measure for liquidity is daily turnover.
To approximate the daily turnover it is simply the Close price times the Volume, and to get a better feel for this price then take a moving average over more than a few weeks and there it is. Very few Technical Analysis programs (apart from Metastock) have the ability to do these calculations, unless if you do them by longhand!
A few other daily average turnover based filters are included to sift out low and very low turnover stocks - to avoid buying into stocks that in reality cannot be sold out of, and this usually leaves less than 300 securities to consider instead of a couple of thousand!.
In general, securities tend to behave differently depending on the turnover value and share price and to further simplify the analysis and selection phase, I have chosen five (5) different security categories, and these are:
With five different pricing and reactionary structures, this leaves about 10 to 60 securities in each type to be considered - instead of about 2000 securities and no structural category.
The associated pages show categorised securities that appear to be in a positive trend - and those that have just appear to have turned to a negative trend in the last week or so. These will automatically sort by clicking on the column heading.
If this does not make sense then tutorials on Technical Analysis are included in this Website.
Copyright © Malcolm Moore, 2002, 2009 Comments and Corrections are welcome