The impact and measurement of the intensity of noise in stock returns

Clements, Adam (2002) The impact and measurement of the intensity of noise in stock returns. PhD thesis, Queensland University of Technology.

Available via Document Delivery only – contact your library to place a request
If you are the author of this thesis, please contact


The development of financial economics literature has been characterised by a continual dialogue between empirical and theoretical researchers. Often, this dialogue has taken the form of empirical observation prompting theoretical enquiry. This thesis follows this long tradition by investigating a number of emerging empirical facts, for which in most cases, simple theoretical explanations are suggested. Broadly speaking, this thesis investigates the manner in which the level of activity in an asset market influences the empirical features exhibited by the asset's returns. Motivated by these empirical observations reported in this thesis, theoretical models based on heterogeneous trader behaviour are suggested as explanations of these observations.

A body of widely accepted empirical facts are first re-evaluated with reference to three representative equity indices. Features such as linear dependence in expected returns, dependence in the volatility of returns and negative correlation between returns and volatility innovations are found to be common characteristics of index returns. A number of authors have documented the emerging fact that the presence of non-linearity in returns is transitory in nature. A central issue of this thesis is to propose a rationale for this as yet unexplained phenomenon. Within a model of trader interaction, it is shown that the intensity of noise trading is critically important for the presence of nonlinear price outcomes. Increases in the intensity of noise trading are shown to extinguish non-linear structure in simulated returns. Analysis of index returns lends support to this notion in that periods of returns that exhibit more intense noise are associated with linearity. Issues relating to the accurate and efficient measurement of noise are discussed in detail. It is found that when dealing with stock returns, simple standard deviation of returns is a valid approximation to the intensity of noise in returns. As the presence of non-linearity in returns does not appear to be a persistent feature, the link between market activity and linear dependence in returns is also investigated. Using a similar model of trader interaction, it is shown that when the rate of news arrival is relatively low (high) strong (weak) positive autocorrelations are detected. Broadly consistent patterns are also detected in index returns, supporting the notion that news influences the behavioural patterns of investors and thus observed structure in returns.

Another emerging empirical fact documented in this thesis is the manner in which the intensity of noise in returns influences dependence in the volatility of returns. An accepted feature of the dependence in volatility is that an asymmetry exists between returns and volatility innovations. It is shown here that during periods where the intensity of noise in returns is relatively high, this asymmetrical effect becomes more pronounced. While no formal explanation of this observation is suggested, this exercise has followed in the tradition of much research in investigating empirical phenomena as a first step in expanding our understanding of asset markets.

The results reported throughout this thesis are important from two perspectives. First, they expand upon our knowledge of the empirical features of asset returns in that emerging facts are re-evaluated and new facts documented. Second, given the theoretical explanations proposed for these observations, insights into the behavioural mechanisms generating returns are also revealed.

Impact and interest:

Search Google Scholar™

Citation counts are sourced monthly from Scopus and Web of Science® citation databases.

These databases contain citations from different subsets of available publications and different time periods and thus the citation count from each is usually different. Some works are not in either database and no count is displayed. Scopus includes citations from articles published in 1996 onwards, and Web of Science® generally from 1980 onwards.

Citations counts from the Google Scholar™ indexing service can be viewed at the linked Google Scholar™ search.

ID Code: 36367
Item Type: QUT Thesis (PhD)
Supervisor: Hurn, Stan
Additional Information: Presented to the School of Economics and Finance, Queensland University of Technology.
Keywords: Stocks Mathematical models, Investments Mathematical models, non-linearity, investor behaviour, hetrogeneous traders, microsimulation market models, noise, conditional volatility, leverage effects, news arrival, thesis, doctoral
Institution: Queensland University of Technology
Copyright Owner: Copyright Adam Clements
Deposited On: 22 Sep 2010 13:05
Last Modified: 08 Mar 2017 05:03

Export: EndNote | Dublin Core | BibTeX

Repository Staff Only: item control page