Cambridge Journal of Economics Advance Access originally published online on May 3, 2005
Cambridge Journal of Economics 2006 30(1):105-122; doi:10.1093/cje/bei049
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Article |
Dealing with the DuhemQuine thesis in financial economics: can causal holism help?
* Dublin City University, Ireland
Address for correspondence: Siobhain McGovern, Dublin City University Business School, Dublin City University, Dublin 9, Ireland; email: siobhan.mcgovern{at}dcu.ie.
Abstract
Acceptance of the DuhemQuine thesis implies a recognition that empirical tests of individual hypotheses are in fact tests of individual hypotheses in conjunction with the networks from which those hypotheses are derived. This leads to a problem in the evaluation of disconfirming evidence: is it the individual hypothesis or the network that is refuted? This paper examines how financial economists deal with this joint-testing problem in their empirical evaluation of the efficient markets hypothesis. Causal holists argue that the holistic testing they espouse circumvents the difficulties surrounding joint-testing problems. This paper assesses whether a causal holist approach can resolve the ambiguity surrounding the evidence against the efficient markets hypothesis. It concludes that, without a resolution of the ontologicalepistemological tension in causal holism identified by Fleetwood (Cambridge Journal of Economics, 2002, vol. 26, 2745) and a clearer outline of the nature of holistic testing, causal holism cannot help financial economists deal with the implications of the DuhemQuine thesis for empirical testing.
Key Words: Causal Holism DuhemQuine thesis Efficient markets hypothesis
JEL classifications: B41, G10
Manuscript received February 17, 2003; final version received December 2, 2004.