On the Network Effect in the Stock Market, The N^3/2 Power Law, and Optimal Volatility Equilibrium

10 Pages Posted: 18 Jan 2010 Last revised: 24 Feb 2010

See all articles by Stephen H.T. Lihn

Stephen H.T. Lihn

Novus Partners, Inc.; Atom Investors LP

Date Written: February 21, 2010


This working paper is an excerpt on the recent finding of the network effect in the stock market. Specifically I summarize the three-half power law which states that the total value of the stock market is proportional to the three-half power of the number of stocks in the market. This power law is based on several intuitive assumptions on a many-body stochastic system, which will be described in this paper. The hypothesis of the optimal volatility equilibrium is introduced as a pillar leading to the power law, which states that such system must adjust itself around a level of optimal volatility in order to survive as a viable market. The connections to Fernholz's diversity index and Clark's lognormal subordinated process are discussed. This hypothesis also indicates that investing in a large market doesn't guarantee the benefit of diversification as prescribed in modern portfolio theory. The tail risk is omnipresent in the financial market.

Keywords: network effect, volatility, lognormal cascade, fat tails, heavy tails, stochastic portfolio theory, diversity

JEL Classification: C22, D3, E32, E37

Suggested Citation

Lihn, Stephen H.T. and Lihn, Stephen H.T., On the Network Effect in the Stock Market, The N^3/2 Power Law, and Optimal Volatility Equilibrium (February 21, 2010). Available at SSRN: https://ssrn.com/abstract=1538274 or http://dx.doi.org/10.2139/ssrn.1538274

Stephen H.T. Lihn (Contact Author)

Atom Investors LP ( email )

3711 S Mopac Expressway
Austin, TX 78746
United States
917-603-4133 (Phone)

Novus Partners, Inc. ( email )

521 5th Ave
29th Floor
New York, NY 10175
United States
917-603-4133 (Phone)

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
PlumX Metrics