There is a wide spectrum of timeframes in which one can invest, from buy and hold for 3 years or more, to a medium term investment lasting 6-9 months, to a 3-month short term investment and all the way down to a week long swing trade or plain intraday trading. Most buy and hold investors realise that the return they have earned is not the best for the period they have invested in it, considering price movement has not been uniformly directional but has had numerous interim corrections, which could have enabled them to earn a lot more if they had called more of these interim corrections correctly. Ideally, they would have like to reverse their position during the interim corrections or at least exit the trade and invest the cash in a savings account or money market fund during the correction. On the other extreme, most day traders ? who try to catch every turn in the market every hour seem to lose money. So if this be the case, there ought to be some point in between where your investment performance could be optimised. What exactly is that time frame? Or in other words what is the optimal trading frequency? But even more critically, should you attempt to optimise your trading frequency? and how should you do it?



The first step in answering these questions is to understand the economics of trading at the points of reversal. Every time a trader exits a position because of an anticipated trend reversal or takes an opposing stance, he obviously incurs costs. These costs can be split into structural costs and analysis costs. Structural costs include brokerage, bid-ask spread (which arises due to illiquidity) and slippage costs which the trader incurs when he cannot get the exact price in fast moving markets. The analysis costs can be broken down into trend identification costs and whipsaws. Unless one was using highly accurate reversal prediction tools (I am not aware of any), a trader would almost always have to wait for some confirmation of reversal by allowing the price to move in the reversed direction for some time. This can be termed as trend identification cost. Even after delaying action to await confirmation of trend reversal, traders are faced with the possibility of a whipsaw or false confirmation of trend reversal. Traders who got out of a long term trend to play a shorter term correction do many times run the risk of missing out on the main long term trend should their assessment of the correction go awry and the stock actually make a big move in the direction of their original position. The return a trader makes during periods of corrections, either by taking a opposite stance or by investing in money markets, should exceed these transaction costs. So the trade-off is between the return earned during the time period for which stock undergoes correction and the transaction costs.

In his book ?Channels and Cycles: A tribute to J M Hurst?, noted technical analyst and author Brian Millard examines the trading frequency issue by studying the 5 year performance of the 30-Dow constituents between 1993 and 1998. Assuming a 5% transaction and illiquidity costs, and assuming that an investor typically enters or exits the trade a few days after the actual trend has begun (the trend identification cost is measured as time delay rather than adverse price in this case), Millard concludes that the optimum trading length should be between 50 and 100 trading days.

Millard's assessment is over ten years old. We have had substantial changes in the markets with the rise (and subsequent crash) of Nasdaq, rapid growth in hedge fund industry and increased individual participation in markets through online trading. With these, I would expect some significant change in Millard's finding. Also, over the last five years, we have been witnessing a steady and steep decline in brokerage costs. By choosing fairly liquid securities with low bid-ask spreads the transaction costs can be brought down substantially. All of these probably would make it meaningful to trade short corrective swings, if one could call the turns correctly. I would have expected the optimum trading length to have come down, to something like 30 to 60 trading days. I am sure one can run a similar study now, based on different input parameters to come out with a more refined assessment but I believe under all combinations of input parameters, there does exist an optimum trading length somewhere between that of the swing trader and the buy and hold investor.

I certainly believe both, the very short term traders and very long term investors would benefit by attempting a minor change in their trading frequency while broadly sticking with their original trading style. The short term trader has to make a conscious attempt at avoiding the excessive transaction costs while the long term investor would be better off anticipating corrections.

Fundamental analysis is never a tool to identify corrections. Playing the corrections requires reasonably fairly accurate timing which fundamental analysis does not promise to identify for you. If you invest solely based on fundamentals, I would recommend that you stick to a buy and hold approach and not attempt to play the corrections. However, technical Analysis can be very effectively used to anticipate trend reversals both in terms of timing and price levels. Adding technical tools to your arsenal can be of immense help.

Similarly, pure technical traders would benefit significantly from getting a fundamental understanding of the stocks that they trade in. A technical trader taking a particular directional position is likely to stick to his original trade even in the wake of a temporary whipsaw, and avoid the stop loss switch, in case he had a certain level of comfort in the immediate fundamentals (or the trend of the next higher order).

There is a wide spectrum of timeframes in which one can invest, from buy and hold for 3 years or more, to a medium term investment lasting 6-9 months, to a 3-month short term investment and all the way down to a week long swing trade or plain intraday trading. Most buy and hold investors realise that the return they have earned is not the best for the period they have invested in it, considering price movement has not been uniformly directional but has had numerous interim corrections, which could have enabled them to earn a lot more if they had called more of these interim corrections correctly. Ideally, they would have like to reverse their position during the interim corrections or at least exit the trade and invest the cash in a savings account or money market fund during the correction. On the other extreme, most day traders ? who try to catch every turn in the market every hour seem to lose money. So if this be the case, there ought to be some point in between where your investment performance could be optimised. What exactly is that time frame? Or in other words what is the optimal trading frequency? But even more critically, should you attempt to optimise your trading frequency? and how should you do it?

The first step in answering these questions is to understand the economics of trading at the points of reversal. Every time a trader exits a position because of an anticipated trend reversal or takes an opposing stance, he obviously incurs costs. These costs can be split into structural costs and analysis costs. Structural costs include brokerage, bid-ask spread (which arises due to illiquidity) and slippage costs which the trader incurs when he cannot get the exact price in fast moving markets. The analysis costs can be broken down into trend identification costs and whipsaws. Unless one was using highly accurate reversal prediction tools (I am not aware of any), a trader would almost always have to wait for some confirmation of reversal by allowing the price to move in the reversed direction for some time. This can be termed as trend identification cost. Even after delaying action to await confirmation of trend reversal, traders are faced with the possibility of a whipsaw or false confirmation of trend reversal. Traders who got out of a long term trend to play a shorter term correction do many times run the risk of missing out on the main long term trend should their assessment of the correction go awry and the stock actually make a big move in the direction of their original position. The return a trader makes during periods of corrections, either by taking a opposite stance or by investing in money markets, should exceed these transaction costs. So the trade-off is between the return earned during the time period for which stock undergoes correction and the transaction costs.

In his book ?Channels and Cycles: A tribute to J M Hurst?, noted technical analyst and author Brian Millard examines the trading frequency issue by studying the 5 year performance of the 30-Dow constituents between 1993 and 1998. Assuming a 5% transaction and illiquidity costs, and assuming that an investor typically enters or exits the trade a few days after the actual trend has begun (the trend identification cost is measured as time delay rather than adverse price in this case), Millard concludes that the optimum trading length should be between 50 and 100 trading days.

Millard's assessment is over ten years old. We have had substantial changes in the markets with the rise (and subsequent crash) of Nasdaq, rapid growth in hedge fund industry and increased individual participation in markets through online trading. With these, I would expect some significant change in Millard's finding. Also, over the last five years, we have been witnessing a steady and steep decline in brokerage costs. By choosing fairly liquid securities with low bid-ask spreads the transaction costs can be brought down substantially. All of these probably would make it meaningful to trade short corrective swings, if one could call the turns correctly. I would have expected the optimum trading length to have come down, to something like 30 to 60 trading days. I am sure one can run a similar study now, based on different input parameters to come out with a more refined assessment but I believe under all combinations of input parameters, there does exist an optimum trading length somewhere between that of the swing trader and the buy and hold investor.

I certainly believe both, the very short term traders and very long term investors would benefit by attempting a minor change in their trading frequency while broadly sticking with their original trading style. The short term trader has to make a conscious attempt at avoiding the excessive transaction costs while the long term investor would be better off anticipating corrections.

Fundamental analysis is never a tool to identify corrections. Playing the corrections requires reasonably fairly accurate timing which fundamental analysis does not promise to identify for you. If you invest solely based on fundamentals, I would recommend that you stick to a buy and hold approach and not attempt to play the corrections. However, technical Analysis can be very effectively used to anticipate trend reversals both in terms of timing and price levels. Adding technical tools to your arsenal can be of immense help.

Similarly, pure technical traders would benefit significantly from getting a fundamental understanding of the stocks that they trade in. A technical trader taking a particular directional position is likely to stick to his original trade even in the wake of a temporary whipsaw, and avoid the stop loss switch, in case he had a certain level of comfort in the immediate fundamentals (or the trend of the next higher order).

<---****HYPERLINK****--->"http://www.submitsuite.com/products/article-submitter.htm">This article was published using Article Submitter