The analytical criteria that we use to identify uptrends and downtrends are tightly defined, and more complex than merely requiring, for example, that the moving averages are rising. Because moving averages lag the movements of actual prices, observation of charts will frequently show that, while the moving averages may be rising, the actual trend is merely moving sideways. Our criteria are designed to filter out data series which are technically in uptrend or downtrend, but which are in reality only drifting sideways.
Here is a full list of the criteria we require for a data series to be formally identified as an uptrend:
• The actual value must lead the 25-day moving average;
• the 25-day moving average must lead the 65-day moving average;
• the 25-day moving average must have been rising for at least 5 days;
• the 65-day moving average must have been rising for at least 1 day;
• to filter out passive uptrends, the data must show a rise of at least 2.5% on the month.
The analytical criteria for a downtrend are exactly the converse of the above criteria.
Any data series not conforming to either set of criteria we describe as having an indeterminate trend They are not listed here. Note that it is quite normal for a data series to alternate between determinate and indeterminate trends from time to time, but impossible for it to jump directly from an uptrend to downtrend, or vice versa without an interval of several days.
The percentage figures which appear in the title box of each section represent the percentage of securities in uptrend or downtrend relative to all securities that we monitor in that section. These figures correspond with the latest uptrend and downtrend percentages shown in the various TrendWatch barometers. We do not require a rise or fall of 2% on the month in order classify the following as uptrends or downtrends:
• FT Actuaries Indices
• World stock markets
This is firstly because there is only a small number of data series in these categories, so it is not necessary to employ a filtering device to separate high and low momentum uptrends. Second, these categories are inherently less volatile than equities. Applying the 2% filter can result in most of the trends being classified as indeterminate, which is undesirable.
The number of securities we monitor varies on a day-to-day basis, as new issues start trading, companies are taken over, warrants expire, etc. We publish a full list of all securities monitored 3 times a year (the beginning of January, May and September. We can send you the latest list on request, for a nominal fee of £3.00 to cover administration and postage.
The shares in these lists carry a suffix indicating the trend duration: i.e. the number of days that the 65-day moving average has been in uptrend or downtrend. Shares in downtrend always carry a negative suffix.
Shares whose 65-day m.a has been in uptrend or downtrend for less than 10 days are prefixed by an asterisk and are sorted in their own alphabetical order at the beginning of the list.
Asterisked shares that fail to become our main recommendations are not necessarily poor buys. They may have been eliminated perhaps because we consider that the positive news is already in the price, or the long-term uptrend is already well advanced or the share may already be in our portfolio.
Please note the following:
• Many shares whose 65-day m.a. is in uptrend or downtrend will not appear in the lists. For a share to get into the list, it must conform with the other trend criteria (see ‘Trend Definition’ overleaf).
• A share may see its 65-day m.a. turn down for a short while and then turn up again. This will result in it appearing as an asterisked entry even though it has only suffered a blip in a long-term trend.
• If the trend becomes indeterminate, a share may disappear from the list and then reappear without being asterisked if the 65-day m.a. remained in uptrend during its absence from the list.
We believe that accurate monitoring of our investment performance is of critical importance, both for you and for us. It is not enough to simply monitor the profit (or loss) on our recommendations. You are entitled to know how we have done relative to the market as a whole. It is no use us boasting of a 20% profit if the market as a whole has risen 30%.
We therefore monitor each of our recommendations against a benchmark index. Ours is the FT All-share Index (exc. investment companies).
Whenever we recommend a share, we record the value of this index as at the date the share was bought. When we do a valuation or when we sell a share, we record the latest value of the index. We then add the percentage change in the index to the cost of buying the share. This tells us how much our investment would have grown had we invested in a market tracker fund rather than in the actual share – the market gain/loss.
To determine how much we have outperformed (or under-performed) the market, it is tempting to subtract the tracker gain from the actual gain. However, this is arithmetically incorrect – you cannot get an accurate figure by performing a subtraction using two percentages. Instead, we use the industry-standard formula:
((100 + actual gain) / (100 + tracker gain)) x 100 - 100
Our performance reporting is scrupulous accurate. For example, if we sell a share at a profit, but the tracker index for that share shows an even bigger rise, we actually record it as a loss against the market. Worse still (for us), if we sell a share at a loss in a rising market, we may record an even bigger loss. But we are doing no more than being honest with ourselves and, more importantly, honest with you.