VARIANCE-RATIO TESTS - SMALL-SAMPLE PROPERTIES WITH AN APPLICATION TOINTERNATIONAL OUTPUT DATA

Two aspects of statistical inference using variance-ratio statistics a re studied, (1) the accuracy of asymptotic approximations in small sam ples and (2) the size distortion arising from searching over many hori zons in deciding whether to reject a model. A joint test combining var iance-ratio statistics at various horizons is proposed, and a Monte Ca rlo procedure for conducting exact inference is provided. The real out put data of nine countries are used to discuss these issues.