Is Technical Analysis Bunk?
Price charts of some sort are almost always used to analyse the financial markets. Every trader from the humble individual to large funds will in some form or other be looking at charts at some stage in their trading. By doing this they are by default using Technical Analysis (TA) of some sort whether they are looking at basic candlestick price charts or RSI or any other indicator.
Therefore, the fact that a set of analyses is so widely used and accepted by the trading community naturally draws a lot of scrutiny as to how useful it is. Just a cursory search on the net will yield many well-argued articles asserting against its usefulness from credible sources such as the Motley Fool and the Financial Times and they may well be right.
In my humble view, this is the wrong question to ask because a trader is free to use anything he wants from reading tea leaves to astrology to fundamental analysis to TA. The only thing that counts when performance is concerned is the bottom line, if a trader shows a strong positive return on capital employed with all the usual caveats in place and he says his success is down to TA or the stars, then who are we to doubt him?
However, this brings into question the issue of objectivity because prices and charts are all made from numbers and numbers which are products of other numbers are then all about objectivity.
Therefore, exactly how will it benefit traders if the charts and indicators used are objective and why do we care if it isn’t? The answer to this is simple, prices are numbers and are products of time which are also numbers (in maths jargon: price on the Y axis and time in the X axis, or the dependent and independent variables respectively). So if a market will exist at any time in the future, then at some stage there will be a price. If so then why should we not use numerical techniques to analyse numbers which produce other numbers (time producing price), in other words why should we not use maths to analyse the markets? It seems obvious to me.
Furthermore I am sure we can agree that the important point in using maths is the objectivity it brings, as it means that charts and indicators should be only created by objective criterion, so that there is only one set of standards determining how the chart or indicator shall be created.
Thus if some line or pattern has to be drawn on our screens then there should be only one way of drawing it, so that we know that its concept is unequivocal and consistent. if not and there are numerous ways of producing it then what it represents will vary and change with the person who draws it. Without a consistent representation then its reliability and utility is of dubious quality. Only consistent and clearly defined techniques will be of value to the trader - this is why the trader should have at least one objective method of drawing and producing charts and indicators.
The example of trendlines will highlight exactly what this means.
TA techniques of dubious foundation
Trendlines and Channels are good old chestnuts of how TA charts that are totally subjective in nature. In classic TA, trendlines are used to confirm a trader’s view of the market, therefore if we are bullish today then we are told to scout around for a market that looks as if it might be going up as in Figure 1.
After this we are supposed to confirm this trend by drawing an upwards sloping line by connecting the lowest low price point to the highest low price as in Figure2.
This is all well and good, but this drawing has no objective criteria because it is subject to the individual’s choice of which points to connect. A very bullish trader could come along and say that he is so bullish that he wants to represent this view by connecting lowest low to the highest high. Another could say that he is not so bullish and only wishes to connect the highest low to the lowest high. Yet another trader could say that his view is in between these two and draw a line right through the middle price points. All these lines still represent an uptrend in some manner, except for the fact that there are no one set of standards for drawing a trendline in that the criteria are completely subject to individual choice. This makes hand drawn lines more art than anything, subjective and not objective (see Figure.3)
The question then is should traders include at least one way to draw the lines based only on objective criteria? And what would that criteria be? This brings us back to the fact that future prices in its purest form are numbers based on other numbers (time mainly) and if we want to analyse these prices then we should use numerical techniques; maths in other words and if we do that then by default we are using objective criteria to analyse prices since maths is almost always an objective process. The maths technique that answers this particular problem is called Linear Regression (LR). LR plots what is known as the best fit straight line through the data by calculating the optimal price on the Y-axis and the optimal angle. The regression trend line can then be drawn straight through these two points. The Y-axis price point and the angle are computed by what is known as the regression matrix.
Fortunately, LR is widely available on most broker software and on Excel and similar software, thus calculating and producing the LR line will be done automatically for you.
Mathemarkets© All rights reserved