finally, this paper highlights the importance of model specification in empirical research. the nonlinear model increases the power of the regression in explaining variation in share price and documents some measurement errors in consolidated data, an inference that would not have been possible had the analysis been confined to the traditional linear model. one possible explanation for such a result could be that, at the first-time adoption of ifrs for consolidated financial statements,fair value was widely applied and many assets were revaluated, which may have induced investor to apply downward corrections when pricing the firm, consistent with the negative coefficient on the nonlinear term.