EVIEWS回回结果的理解(国外英语资料).doc

EVIEWS回归结果的理解 The understanding of regression results Parameters: Regression coefficient (coefficient) Note that the positive and negative of the regression coefficients should conform to the theory and practice. The regression coefficient of the intercept term is no real economic significance whether or not it is passed by T test. The standard Error of regression coefficient (Std. Error) The greater the standard error, the less reliable the estimate of the regression coefficients, which can be calculated by the formula of T T - Statistic T test is whether the regression coefficient is equal to a certain value, the specific value of 0 in the regression equation, so T value = / regression coefficient of standard error of regression coefficient, so T value of positive and negative should be consistent with the regression coefficient of plus or minus, standard error of regression coefficient, the greater the T value is smaller, the estimate of regression coefficients is unreliable, the more close to zero. In addition, the greater the absolute value of the regression coefficient, the greater the absolute value of T. 4, P values (Prob) P value is the probability of the T value theory T value beyond the sample, should contact significance level than alpha, alpha said under the premise of the original hypothesis, sample theory T value than the probability of the T value, when the P value alpha value, indicating the result actual appearance probability of probability than in the original hypothesis under the premise of the possibility of the results is small but it appeared, so refused to accept the null hypothesis. (r-squared) It is known that the coefficient of determination means that the interpretation of the variable is explained by the interpretation of the variable, which is the degree of consistency between the y and the y equals y. The y value of y is the population mean of y. Adjusted refraction (Adjusted r-squared) Namely, the degree of freedom adjusted r-squar