Unpredictable market? Per aspera ad astra


As soon as the stock market has played an important role not only in the economy but also in politics and social life, began to watch him not only neposredstvennye members, but also people who have no direct relation to the market. The bankruptcy of one of the first notorious <pyramids> - West India Company in France - and the stock crash of the late 20's of the last century, which began with the Great Depression, the stock market made a focus of attention of the general public. Therefore projections for him as a tool <struggle> with the uncertainty of the future is becoming increasingly important. This information could influence the political decision-making or trigger a wave of social unrest. It is not surprising that the issue of predictability of the market attracted the attention of serious scientists. With their usual thoroughness and breadth posed the question of whether the market is predictable and, in principle?

The classical period

On reasoning about it initially had a strong influence of the known physics of Brownian motion and the central limit theorem. English physicist Brown discovered a remarkable phenomenon: if a small crumb of paint put into a liquid, it does not remain in place, and begins to move erratically, and traffic increases with increasing temperature.

Explanation has been given the following: a liquid composed of invisible small particles, which are in constant motion. We can not watch it due to the size of the particles, but we see the movement of small particles of paint, resulting in millions of collisions between it and the fluid particles. Since the particles in the fluid are moving independently of each other, their number is extremely large and they are all the same, collisions occur randomly, and the total momentum transfer is also a random quantity distributed, due to the central limit theorem, the normal distribution. This explanation is not to predict the exact movement of specific particles, but it allows us to obtain statistical estimates of its trajectory, finding the mean and variance of the displacement relative to the initial position.

Because these arguments were effective for molecular physics, an attempt was made to use them to explain the dynamics of the stock market. Test particle there is a specific asset or market index, the location of one-dimensional motion - its price, liquid particles - the market participants. As a result - the price is a random walk, which is the Brownian motion.

External similarity of real historical data and artificial time series, modeled under the assumptions made, predetermined confidence of researchers in this explanation of price and spawned the efficient market hypothesis. If we translate the requirements under which the applicable unit of Brownian motion, with the mathematical language in home, you get the following:

- Market participants must act independently of each other;

- The number of bidders for each asset must be extremely large;