The recent post suggested that a variety of market observations can be pulled together by drawing upon an explanatory framework. A good theory not only helps us understand what has happened in markets; it also suggests what we might observe in the future. Science begins with observation. Until we link our observations and make sense of them, they remain isolated data points. Theory is a big part of what transforms data into explanation.
What my observations suggest is that a limited number of variables, such as buying and selling pressure; volatility; breadth; sentiment; and correlation, uniquely predict short-term price movement in the stock market. An ongoing research interest has focused on the structure of cycles in the stock market and the co-movement of these variables at different phases of market cycles. In general, I find that market cycles can be broken down into several phases:
1) Bottoming process - At market lows, we tend to see an elevation of volume and volatility and a high level of market correlation, as stocks are dumped across the board. Selling pressure far exceeds buying pressure and sentiment becomes quite bearish. At important market bottoms, we see price lows that are not confirmed by market breadth, as strong stocks begin to diverge from the pack and attract buying interest. At those bottoms, we also find a rise in buying pressure and a reduction of selling pressure, as fresh market lows fail to attract new selling interest.
2) Market rise - With the drying up of selling, low prices attract buying from longer timeframe participants as well as shorter-term opportunistic ones. The market rises on strong buying pressure and low selling pressure, and the rise generates sufficient thrust to generate a good degree of upside momentum. Volatility and correlation remain relatively high during the initial lift off from the lows and breadth is strong. Dips are bought and the rise is sustained.
3) Topping process - The market hits a momentum peak, often identifiable by a peak in the number of shares registering fresh highs. Selling from this peak generally exceeds the level of selling seen during the market rise, but ultimately attracts buyers. Weak stocks begin to diverge from the pack and fresh price highs typically occur with breadth divergences and lower levels of correlation. New buying lacks the thrust of the earlier move from the lows and volatility wanes. By the time we hit a price peak for the cycle, divergences are clear, volatility is low, both buying pressure and selling pressure are low, and sentiment remains bullish.
4) Market decline - Fresh selling creates a pickup in correlation and volatility, as short-term support levels are violated and selling pressure exceeds buying pressure. Breadth turns negative and the bulk of stocks now move lower.
The thorniest problem I have encountered in my work with markets concerns the timing relationships among these phases of market cycles. I am convinced that the cycles are aperiodic (they do not adhere to rigid timetables; there is no invariant x-day cycle), and I am also convinced that there is a non-random proportionality among the phases that occurs within and across time frames. Capturing this proportionality has been the greatest challenge in the research.
The nesting of larger and smaller cycles creates periods of apparent trending and periods of apparent range-boundedness in markets. Optimal trading strategies conform to the parameters of the cycles operating in markets at a given time. Problems occur in trading when a trader's holding period greatly differs from the operative cycle period(s). A common problem in trading today's stock market is that the operative cycles are much longer than most active traders' time horizons. This leads to chasing nickles in front of larger move steamrollers.
By tracking the ongoing changes in market sentiment, volatility, breadth, correlation, and buying/selling pressure, we can identify where we stand in a market cycle and adjust trading strategies accordingly. During market rises and declines, we want to be holding positions and going with momentum; during topping and bottoming processes, we want to be fading range extremes and trading more tactically. In general, we want to be aligned with the largest operative cycles, as these greatly impact the shape and timing of smaller cycles.
I am mindful that it is easy to find "cycles" in hindsight, with little explanatory benefit. The cycle concept only has value insofar as it organizes observations among key market variables and suggests observations to come. I will be elaborating the explanatory and predictive value of cycles in future posts; it is the major focus of my market research. For an excellent example of cycle research creating profitable trading strategies, check out the StockSpotter site.
Further Reading: Market Profile as a Practical Theory
.
What my observations suggest is that a limited number of variables, such as buying and selling pressure; volatility; breadth; sentiment; and correlation, uniquely predict short-term price movement in the stock market. An ongoing research interest has focused on the structure of cycles in the stock market and the co-movement of these variables at different phases of market cycles. In general, I find that market cycles can be broken down into several phases:
1) Bottoming process - At market lows, we tend to see an elevation of volume and volatility and a high level of market correlation, as stocks are dumped across the board. Selling pressure far exceeds buying pressure and sentiment becomes quite bearish. At important market bottoms, we see price lows that are not confirmed by market breadth, as strong stocks begin to diverge from the pack and attract buying interest. At those bottoms, we also find a rise in buying pressure and a reduction of selling pressure, as fresh market lows fail to attract new selling interest.
2) Market rise - With the drying up of selling, low prices attract buying from longer timeframe participants as well as shorter-term opportunistic ones. The market rises on strong buying pressure and low selling pressure, and the rise generates sufficient thrust to generate a good degree of upside momentum. Volatility and correlation remain relatively high during the initial lift off from the lows and breadth is strong. Dips are bought and the rise is sustained.
3) Topping process - The market hits a momentum peak, often identifiable by a peak in the number of shares registering fresh highs. Selling from this peak generally exceeds the level of selling seen during the market rise, but ultimately attracts buyers. Weak stocks begin to diverge from the pack and fresh price highs typically occur with breadth divergences and lower levels of correlation. New buying lacks the thrust of the earlier move from the lows and volatility wanes. By the time we hit a price peak for the cycle, divergences are clear, volatility is low, both buying pressure and selling pressure are low, and sentiment remains bullish.
4) Market decline - Fresh selling creates a pickup in correlation and volatility, as short-term support levels are violated and selling pressure exceeds buying pressure. Breadth turns negative and the bulk of stocks now move lower.
The thorniest problem I have encountered in my work with markets concerns the timing relationships among these phases of market cycles. I am convinced that the cycles are aperiodic (they do not adhere to rigid timetables; there is no invariant x-day cycle), and I am also convinced that there is a non-random proportionality among the phases that occurs within and across time frames. Capturing this proportionality has been the greatest challenge in the research.
The nesting of larger and smaller cycles creates periods of apparent trending and periods of apparent range-boundedness in markets. Optimal trading strategies conform to the parameters of the cycles operating in markets at a given time. Problems occur in trading when a trader's holding period greatly differs from the operative cycle period(s). A common problem in trading today's stock market is that the operative cycles are much longer than most active traders' time horizons. This leads to chasing nickles in front of larger move steamrollers.
By tracking the ongoing changes in market sentiment, volatility, breadth, correlation, and buying/selling pressure, we can identify where we stand in a market cycle and adjust trading strategies accordingly. During market rises and declines, we want to be holding positions and going with momentum; during topping and bottoming processes, we want to be fading range extremes and trading more tactically. In general, we want to be aligned with the largest operative cycles, as these greatly impact the shape and timing of smaller cycles.
I am mindful that it is easy to find "cycles" in hindsight, with little explanatory benefit. The cycle concept only has value insofar as it organizes observations among key market variables and suggests observations to come. I will be elaborating the explanatory and predictive value of cycles in future posts; it is the major focus of my market research. For an excellent example of cycle research creating profitable trading strategies, check out the StockSpotter site.
Further Reading: Market Profile as a Practical Theory
.