A long-standing conventional view of the mainstream economists to the financial markets is expressed in an Efficient Market Hypothesis (EMH), which views asset prices and their associated returns from the perspective of the speculator, and assumes the market is almost always efficient, meaning that the prices already reflected all current information that could help anticipating future events.

Under the EMH, the stochastic process of market returns is modeled as uncorrelated random walk, so any profitable prediction of the market returns would be considered impossible. Although the strong form of EMH has been shown not true by many studies as reviewed by Pan (2003), EMH does provide a perfect reference with which more realistic market views can be developed.

We consider that the market is driven by various forces of different origins, wave lengths and magnitudes, as well as different durations. Each force only has one dimension: either positive for upwards or negative for downwards. At any given time, all the effective forces combine to form a joint force which has a joint market impact leading to changes in price (or index).

For generality, we will just talk about the dynamics of the price of a market. If the market is a stock, the price of the market is just the price of that stock. If the market is the whole stock market for a national economy, the price of the market refers to a benchmark index such as ASX S&P 200 index for the Australian stock market.

Following Mandelbrot’s discovery of fractals in the financial markets, Peters (1994) proposed a Fractal Market Hypothesis (FMH) which basically says that the market is stable when it consists of investor covering a large number of investment horizons, and information is valued according to the investment horizon of the investor. Because the different investment horizons value information different

ly, the diffusion of information will also be uneven. At any one time, prices may not reflect all available information, but only the information important to that investment horizon (or time frame in our terms). The FHM owes much to the Coherent Market Hypothesis (CMH) of Vaga (1991) and the K-Z model of Larrain (1991).

Like the FMH and CMH, our view to the market, which we call the Swing Market Hypothesis (SMH), is also based on the premise that the market assumes different states and can shift between stable and unstable regimes. However, we take one step further in considering the dynamic structure of the market. The Swing Market Hypothesis (SMH) proposes the following:

(1) The market always consists of investors or traders with all possible different capitals, different time frames, different information conditions, and different skills. The everlasting and ever-evolving differences in investors or traders are the permanent drivers of market dynamics.

(2) The market is sometimes efficient and other times inefficient, and the market tends to swing between these two modes intermittently. Each mode, efficient or inefficient, may comprise multiple regimes such as trending, cycling, spiking, consolidation, etc.

(3) The market movement is driven mainly by three types of forces: business dynamics, mass psychological dynamics, and new impacts.

(4) The market movement can be decomposed into four types of components: dynamical swings, physical cycles, abrupt momentums and random walks.

Business dynamics include global and national business cycles, intramarket dynamics and intermarket dynamics, which usually refer to the fundamental economic and business conditions. Mass psychological dynamics include the greed and fear dynamics of investors and traders defined by human nature and also forged by existing popular knowledge, methodologies and technologies of technical analysis and fundamental analysis.

For example, the appearance of a certain chart pattern may trigger similar trading decisions made by many different technical traders because they have all acquired similar technical analysis education. Both business dynamics and mass psychological dynamics can produce similar dynamical swings which can be characterized by mathematical fractal models such as log-periodic power laws. Elliott waves of different levels are the visual and qualitative description of dynamical swings by technical analysts. Although not as strong as dynamical swings, physical cycles do exist in the market, which includes anniversary days, seasonality cycles and weekly cycles.

For example, statistical studies show that Thursday of a week is often the reversal if the first three days have trended in the same direction. There are even intraday dynamical patterns given the daily market situation. Physical cycles have relatively constant periodicity. Abrupt momentums refer to drastic price movement which cannot be expressed in continuous analytical forms. Momentums may be caused by exogenous forces such as news shocks or impacts, or as critical points or singularities caused by endogenous dynamics. Random walks in the context of SMH correspond to remaining randomness not explicable by any systematic force.

The next section derives a simple but fundamental price impact model of the stock market returns driven by market forces.

Prof. Heping Pan

Next: A Fundamental Price Impact Model of The Stock Market

Summary: Index