Oliver Linton is a Professor of Political Economy and Econometrics at Cambridge University and a Fellow of Trinity College. He has served as an Associate Editor with Econometrica, a co-editor at Journal of Econometrics, Econometric Theory, and a joint editor of Royal Economic Society (RES) Econometrics Journal. Linton is a Fellow of the British Academy, a Fellow of the Econometric Society, and a Fellow of the Institute of Mathematical Statistics. His research contribution has mostly been to do with nonparametric and semiparametric methods. He is also interested in Financial Econometrics.

We introduce a general testing procedure to discriminate two broad classes of econometric models of asset prices. Under the null hypothesis that grounded in no-arbitrage, the asset prices evolve as a semimartingale that exhibits stochastic volatility and jumps; under the alternative, a pricing error that reflects various market frictions is added to the semimartingale part. Using high-frequency data, we derive limit theorems for the proposed testing statistic when the asset price follows a general semimartingale, leading to formal tests for the presence of pricing errors. Our test has power against a broad class of alternative hypotheses where the pricing errors could be endogenous, nonstationary, seasonal and serially correlated. Moreover, the observation times are allowed to be stochastic under both the null and the alternative hypotheses. The testing procedure is uniformly valid under both the infill (with fixed time span) and non-infill, long span asymptotic frameworks. It thus finds broad applications in time series analysis on market microstructure, asset pricing and macroeconomics where the underlying time series has a permanent and a transitory component.