Non Random Behavior in Financial Markets
54 Pages Posted: 24 Jan 2018 Last revised: 25 Nov 2018
Date Written: January 17, 2018
What is the nature of the price formation process? Is it purely random or not? If so, does that mean inefficiency? With this paper we attempt to answer those questions by providing evidence of a structural, non-random, predictable behavior in financial markets, for all asset classes, at any frequency, proposing a theoretical framework in which such evidence can coexist with a broader efficient market hypothesis. We treat the price formation process as a mixture of patterns, or two-dimensional objects, made by a non-divisible combination of price changes over time changes, where both quantities are random variables. In this context, the analysis of returns becomes an overly simplistic measure to understand market behavior, since it alters the object by imposing a deterministic structure to its time dimension, linked to the data generating process. By developing a methodology that is able to extract trend lines from prices and identify their breakout moments, we try to capture and analyze the full price-time dimensional structure of the most elementary price pattern, i.e. market trends. We show that strong memory in trends can coexist with absence of memory in returns over the same time series of prices, separating de facto the concept of efficiency from the concept of randomness. Consequently, the autocorrelation of returns may be considered a proof of market efficiency only and strictly under the agents’ rationality assumption, but it cannot be considered a proof of market randomness or unpredictability. Finally we show, through the implementation of a trading strategy, how the evidence found can be used to obtain systematic extra returns on any asset class, at any frequency, suggesting the presence of a universal, structural, intrinsic behavior in financial markets.
Keywords: Prices, Returns, Randomness, Market Efficiency, Behavioral Finance, Market Behavior, Autocorrelation of Returns, Technical Analysis, Pattern Recognition, Asset Allocation, Trading, Fractals
Suggested Citation: Suggested Citation