Endogenous parameter time series estimation of the Ohlson model: Linear and nonlinear analyses

This paper tests the empirical content of Ohlson's model in three ways. First, abstracting from the nonlinear relationships among the parameters, the estimated coefficient signs and values are evaluated relative to the prediction of the Ohlson model. Second, recognizing the underlying non-linear relationships among the parameters, the Ohlson model is evaluated for internal consistency by comparing parameter estimates from the earnings and valuation equations. For internal consistency to hold, the two equations should yield parameter estimates that are not significantly different from each other. Third, the model is estimated as a system of restricted nonlinear equations where it is assumed that the parameters are equal to each other across equations. The resulting parameter estimates are evaluated for statistical significance.