Randomization, Endogeneity and Laboratory Experiments

John Ham, John Kagel and Steven Lehrer

No 1524, Econometric Society World Congress 2000 Contributed Papers from Econometric Society

Abstract: In conducting experiments with multiple trials, outcomes from previous trials can impact on current behavior. One of the most obvious cases in which this can happen, and the case considered in this paper, is in an auction market experiment, where earnings from previous auction trials alter cash balances which, in turn, can affect bidding behavior. (The most obvious mechanism for such a result, within standard theory, is if bidders are risk averse and do not have constant absolute risk aversion. One can imagine a number of non-standard reasons for such effects as well.) Use of OLS regressions with cash balances included as a right hand side variable are likely to lead to a biased estimate of the cash balance effect since the variation in cash balances is largely related to differences in bidding strategies across individuals. Fixed effect regressions can commonly control for these endogeniety problems at the potential cost of obtaining inefficient estimates, since this estimator does not exploit between-individual variation. This paper addresses this problem in two ways. First we consider an experimental design that reduces the potential bias of OLS estimates while increasing the precision of fixed effect estimates. Second, we consider instrumental variables estimation of the cash balance effect where the instruments are produced by the experimental design. To the best of our knowledge, neither of these approaches has been explored in the experimental literature.

Date: 2000-08-01
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (7)

Downloads: (external link)
http://fmwww.bc.edu/RePEc/es2000/1524.pdf main text (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ecm:wc2000:1524

Access Statistics for this paper

More papers in Econometric Society World Congress 2000 Contributed Papers from Econometric Society Contact information at EDIRC.
Bibliographic data for series maintained by Christopher F. Baum ().