Department of Finance and Cooperative Management National Taipei University;Department of International Business, National Taiwan University, Taipei, Taiwan
The prior studies note that the floating-point exception (FPE) in computing may lead to under-estimation of the probability of informed trading (PIN), which interests market microstructure empiricists. This study further finds that the FPE bias may result in the underestimated coefficients of market beta in the regression for assets pricing. Namely, the FPE bias may contaminate the tests for pricing, and should be eliminated in the further studies. After eliminating the FPE bias, we document that the PIN factor appear to be a significant factor only during the period of medium level of market liquidity.