Pourquoi l’hypothèse de marche aléatoire en théorie financière ? Les raisons historiques d’un choix éthique

Abstract

The hypothesis of random walk in financial theory : some historical arguments to explain an ethic choice This paper analyses the way modern financial theory originates in the development of financial markets in France during the mid-19th century. We show that the random walk model with a normal distribution was built by Jules Regnault in 1863 - a model to be found in Bachelier (1900) -, at a time when the development of financial markets gave rise to new questions about the morality of these markets. More explicitly, the choice for Gaussian distribution as well as for random walk model originated in ethic and not in empirical considerations. If the efficient theory, built in 1970 by Fama, presents this choice as an informational problem, this paper suggests that our current science schemes are still partially dependent on 19th century views. JEL classifications : B10, A12, G20

Econophysics: A New Challenge for Financial Economics?

“…This generalization is based, on one hand, on Lévy's work (1924) on random processes, and, on the other hand, on the generalized central-limit theorem developed by Gnedenko and Kolmogorov ( 1954 ). In accordance 51 Jules Regnault, in 1863, was directly infl uenced by Adolphe Quételet's work on the application of normal distribution to social phenomena (Jovanovic 2001(Jovanovic , 2006. Bachelier ( 1900 ), whose work was clearly infl uenced by Regnault's (Jovanovic 2000(Jovanovic , 2009b(Jovanovic , 2012, retained a Gaussian description of the evolution of variation in the price of assets for demonstrating the equivalence between the results obtained in discrete time and in continuous time.…”

Was There a “Vernacular Science of Financial Markets” in France during the Nineteenth Century? A Comment on Preda's “Informative Prices, Rational Investors”

“…Unlike other publications, which aimed at educating investors, Regnault's book contains theoretical models and empirical tests, 4 and it took its place nicely in the academic and scientific debates of the time about the social sciences-it used in particular conventional scientific criteria (i.e., criteria of academic science). More precisely, this book extended Adolphe Quételet's research program to a new field: the determination of "scientific laws" that rule financial markets (Jovanovic 2001(Jovanovic , 2006a. 5 We know that Quételet was the first to realize the Condorcet and Laplace program, which aimed at applying the theory of probability to the social world.…”

The History of Econophysics’ Emergence: A New Approach in Modern Financial Theory

“…Both are recent developments -less than 50 years old. While some studies on what was to become modern financial theory were produced prior to the 1960s (Poitras 2000, Preda 2001, Courtault and Kabanov 2002, Dimand 2004, Preda 2004, Jovanovic 2006a, Poitras 2006, Poitras and Jovanovic 2007, they were marginal 1 and did not yet constitute either an academic or a scientific discipline; applied mathematics and empirical investigations into finance existed, but these were isolated contributions, and most of them did not have a solid theoretical underpinning 2 .…”

“…Theoreticians pointed out the absence of theoretical explanations during the 1950s. This was particularly striking after the Koopmans-Vining debate in the late 1940s, which set the NBER against the Cowles Commission over the lack of theoretical explanations and the need to link measurement with theory (Jovanovic 2008). history of econophysics has been produced 3 .…”

“…Two 3 The few papers that present econophysics (Yegorov 2007, Săvoiu and Iorga-Simăn 2008, Daniel and Sornette 2010 provide no exhaustive historical analysis of the approach. 4 As Jovanovic (2008) and Jovanovic and Schinckus (2010) explain, the major hypotheses, models and results of financial economics find their origin in modern probability theory; its institutionalization was also made possible by the link with modern probability theory. crucial elements are identified: first, the evolution of statistical and probabilistic tools; second, the emergence of new empirical data.…”