Daryl Guppy is founder and Director of Guppytraders.com. He is a full-time active private position trader, trading equities and associated derivatives markets. He is an appointed foundation member of the Australian Government Shareholders and Investors Advisory Council.
He is the developer of the Guppy Multiple Moving Average Indicator included in EzyChart. He delivers accredited courses for the Singapore Stock Exchange. He also runs public trading workshops, and equity and futures brokerage sponsored seminars, throughout Australia and Asia.
USING A PRIVATE INDEX
A private index is a tool for monitoring the effectiveness of your selected trading techniques. It is sometimes confused with a portfolio tracking tool which tracks the success of individual trades. The private index helps you to understand how well you are doing in relation to the market. Are you in touch with the market, or are you losing touch? Is the failed trade part of a pattern, or just a normal part of the balance of probabilities in trading? The private index helps to answer these questions.
Usually, at the end of any selection process, we have to make an intuitive choice between two or more equally valid trading candidates. Luck sometimes plays a role and we select the best, or worst, performing of these three candidates. When the trade fails, we may assign the result to bad luck, but this failure may mask a developing failure in our selection technique. When a trade succeeds, we assume this is due to our skill, but even this may be due to luck and mask a failure in our selection techniques. The Private index provides a way to monitor the ongoing success of our selection and analysis process.
The starting point for a private index is our personal selection processes. For simplicity, we assume traders have just a single selection process. In reality, traders often have several distinct selection methods and these can all be tracked with individual private indices. A selection process is a consistent series of steps used to identify and select stocks for further analysis. We could use a technical search that selects stocks where the 7 day moving average has crossed above the 21 day moving average sometime in the past 5 days. Filter these results to eliminate those where the degree of separation between the averages is less than 3% or more than 8%. Further filter the results to remove stocks where the average 10 day value of trading has been less than $100,000 per day. Finally, focus only on those stocks where price is between $0.40 and $1.50.
This is not a particularly complex search technique, but like all technical searches, it reflects our belief about the market and opportunity. This search tells us that the trader believes the market offers trending opportunities in stocks with consistent liquidity and that these opportunities are best exploited using a reasonable level of price leverage. Further, this search is interested in finding newly established breakout opportunities as a new uptrend emerges. Armed with this approach, the trader searches for just this particular type of opportunity, because he believes this is the best way to trade the market at this time.
But is it? Most times we track the success of our selected approach by looking at the individual trades we actually take based on this search. However, this skips a very important step. At the end of this search, the trader may have a list of 30 candidates, particularly if his selected search is compatible with current market behaviour. From this list, he may narrow the choice down to perhaps 10 candidates, all of which are virtually indistinguishable. We call them finalists. They all look equally good, so how do we choose between the finalists? We simply give our biases free range, rejecting stocks in sectors we think will not perform, cutting those we have read bad reports about, ignoring those where we think the price is too high. The final selection is a very intuitive process based on guesswork, instinct, gut feel and luck. It is often an ignominious end to a process that has been carefully controlled up until this point. It is relatively acceptable because the stocks we are choosing between are all excellent opportunities. We have to make a single choice because we do not have enough capital to trade all the choices.
Once we have selected a stock to trade, we usually forget the other finalists that were on the list. These forgotten stocks tell us important things about the overall success of our preferred selection method, and by implication, about our understanding of the market. We compile this information using a personal or private index that tracks the ongoing performance of the stocks that made it to the final selection list. In this example, these are the 10 finalists.
If our selection and stock screening process is valid, then we expect it to consistently turn up winners. These stocks, as a group, should out-perform the market. If they do not out-perform the market, then our selection process is flawed. If these 10 finalists under-perform the market, then our understanding of the market is not appropriate. We are simply out of touch, and although some individual trades may be winners, on balance our selection process is going to deliver a list of losers.
The starting point for comparison is the general market index. We use the XJO as the benchmark because the selection process described above searched for any stocks within the entire market that met the conditions. If the search was limited to a particular sub-sector, then that sub sector index is the appropriate benchmark. In all cases, our objective is to verify that our selection process is capable of outperforming the market.
The market index is an ongoing, extended, continuous calculation that extends back for many years. Our selection process does not have this longevity and this raises important questions about how we apply a private index. The private index is not an exercise in back testing. We are evaluating current stock selections - the 10 finalists - against the current performance of the market. Our intention is to evaluate these selections from the time they were made to the time when the selection process is no longer valid.
Our sample selection looked for stocks that had recently broken into an uptrend. Previously, they were in a downtrend. If we used previous price data while they were in a downtrend, then as a group they would under-perform a rising market. Such information is irrelevant to our task because these stocks were not selected while they were in a downtrend. Our purpose in the private index is not to assess the performance of an individual stock, or the long term performance of a trading system based on a 7 and 21 day moving average.
Our objective is to assess the success of our current stock selections in comparison to the current market. The construction of the private index starts on the day on which the 10 finalists were selected. For this example, we use August 1 and call this PI2. The value of this index continues to be calculated every day until we stop using this particular selection method. The index tracks the performance of our selection method by aggregating the performance of the top 10 finalists. This requires closer management than we usually apply to the finalists who did not actually make it into a trade based with cash.
The market index is adjusted regularly, with stocks added and others dropped. We do the same with the private index. The 10 finalists that initially make up PI2 are paper traded using the same management techniques we apply to the actual trade based on this selection technique. If the trading plan calls for an exit when the 7 day moving average crosses below the 21 day moving average, then this is applied to the 10 finalists. When these conditions are met, the stock is dropped from the index. The same applies when a stop loss condition is met. The result is a private index that reflects not only the choices made, but the ultimate success or failure of those choices. After all, once the new uptrend ends and turns to a downtrend, we would not want to trade the stock, so its performance should no longer be included in the private index calculations.
This is a dynamic calculation. When the real trade is closed, we may choose to repeat the search, again looking for suitable trading candidates. The performance of the private index will tell us if it is worthwhile using the same search and selection method. Even if the individual real trade was not particularly successful, the selection process captured and tracked in the private index may still be outperforming the market. In this case, we can use the selection process with confidence. If the selection process, as tracked by the private index, is under performing the market, then we know we need to re-assess the way we search for opportunity.
When the search and selection is repeated, the 'finalists' are also added to the private index. At this point, the index may consist of several stocks from the initial search which have not yet triggered exit signals, and the stock finalists from the new search. Potentially, this index can expand to include 20 or 30 stocks, each of which is paper traded and closed based on the trading plan conditions applied to the genuine open trade.
If we use several distinct analytical approaches to the market, then we should use a separate private index for each. This is particularly useful if we want to test a particular approach because we believe market conditions are changing. The private index will tell us when the new approach is an effective solution to the change in the market. This sounds like hard work, but it is both necessary and less work than expected. Private Index construction can be achieved in three ways.
Trading offers many challenges. Of these, the most significant is the trader's ability to remain in-touch with the market. Although there are times when the market appears to set an established and long lasting pattern of behaviour, there are always subtle changes developing. Long term success in the market demands that we recognise these changes and adjust our trading strategies accordingly. The analysis and search techniques that turn up successful trading opportunities in one phase of the market become less successful as the market changes. In extreme cases, they become completely inappropriate and the first we know of it is when trades start to fail.
Here is the danger. The trade fails. We accept that this is consistent with the average rate of success. We expect some failures. After a string of successful trades, we can accept a string of unsuccessful trades because the overall average of trading success remains above 70%. Sometimes the expected 30% of losing trades do clump together, one after another. But what if this clumping occurs because we have lost touch with the market conditions? Now it is our initial selection process that needs to be changed.
If we focus just on the performance of our open trades, we miss the forewarning provided by tracking the general performance of the finalists found using our preferred scanning and screening process. If our open trades start to fail, and the finalist trades also fail, then we are looking at a broad failure of our selection processes. We are losing touch with the market and it is time to change tactics and analysis methods.
The private index tracks the ongoing performance of all selection choices you intend to trade. These are the finalists. The combined success, or failure, of this group is an index of the success or failure of our chosen selection method. If we know our selection method is generally outperforming the market in the current period, then we can use the same selection method to find new opportunities as capital becomes available. The private index gives traders a more complete picture of their ability to stay in touch with the market and changes in the market.
First Published: 30 November 2009 - Copyright © Daryl Guppy
DISCLAIMER AND COPYRIGHT