16 July 2010
Trading Moving Averages
There are a great many column inches devoted to the topic of ‘trading’ as opposed to investing. PSG’s Mark Seymour recently wrote a technical piece on trading moving averages that helped to illuminate this-
‘There is a simplistic beauty in following an investing system such as buying & selling solely on a moving average. Buy when the price goes above the moving average - Sell when the price drops below the moving average. The theory is: Capture the up-trend of the market and cutting one’s losses when the market corrects - Outperformance ... right?’
The moving average is the average price of an asset, such as an equity, over a defined period of time (e.g. 7 days, 50 days, 200 days etc). By updating this average each day a daily instruction to0 buy or sell can be generated:
‘For arguments sake let’s take a look at how the 200-day trading strategy has worked for the All Share Index (assuming dividends reinvested). We will assume the following: 1) We buy when the All Share Index goes above the 200-day moving average and sell when the index drops below the 200-day moving average (see figure 1 below), 2) We achieve money market returns when we switch out of equities, 3) After the index crosses the moving average, we get the following day’s price (i.e. 1 day reaction time), 4) No switching fees and 5) Ignore capital gains-tax implications.
Figure 1 - All Share Index (dividends reinvested) & 200-day Moving Average
Source: I-Net Bridge
Results
Amazingly this strategy has worked relatively well over the period July 1997 to the present (see figure 2 below). The trading strategy would have yielded the same returns as the All Share Index with arguably less severe draw-downs.’
The All share Index is a South African index but Mark’s hypothesis, with different data applies equally well to many equity indices
Figure 2 - All Share Index (TR) vs. 200-day moving average trading strategy
Source: I-Net Bridge
However theory and practice aren’t always the same-
“When it comes to practical implementation of this strategy, however, certain complications arise. Let’s focus on two initial assumptions which have a huge impact on returns. These include “number of days used in the moving average” and secondly the “no cost” switching assumption. By using different numbers of days in the moving average calculation leads to a change in the number of switches one has to implement which impacts on the return dynamic. Secondly switching carries a cost which severely hampers returns.
Results taking into account practical complications
Table 1 - Results for 200-day MA and changing costs
Source: I-Net Bridge
There’s also the problem of choosing the right moving average something that many only be apparent is hindsight –
Table 2 - Results for fixed 0.25% cost and changing moving average
Source: I-Net Bridge
Based on the results above one gets an appreciation for the huge impact costs have on returns. High costs leads to severe under-performance. Secondly, if the costs were limited to
0.25% per switch it indicates an optimal number of days one should use for the moving average. Too short a moving average leads to large number of trades which pushes up total cost. Too long a moving average leads to slow reaction time for switching out of the market leading to market-like draw-downs and overall under-performance.
Closing remarks
Although this study suffers from sample selection bias in that we have analysed only one data-set over one particular period, two important lessons can still be learnt. These are: 1) Even seemingly low costs have a large negative impact on returns regardless of investment strategy, 2) Changing an element in the investment strategy (in this case changing the number of days in the moving average) has a huge impact on returns.
Before embarking on an investment journey it is therefore worth understanding the costs involved, what strategy will be followed and how consistent the process will be applied. Lastly, there seems to be no case for following a moving average based strategy as described above, despite its simplistic beauty.”