● Pr>|T| = p-value for a two-tailed t-test of whether the population average is zero. Lowp-values provide evidence that the population mean is not zero. ( Incidentally, you willprobably never see SAS label a p-value by that name. Instead, SAS will phrase the pvalueas a probability. In this case, Pr>|T| indicates the probability of achieving a largerabsolute value of the t-test statistic if, indeed, the population average is zero.)● Num ^= 0 is the number of observations not equal to zero.● Num > 0 is the number of observations greater than zero.● M(Sign) = (number of >0 elements) - (number of non-zero elements)/2, to test that themedian is 0.● Pr>=|M| is the p-value for the two-tailed sign test.● Sgn Rank = Wilcoxon signed rank statistic for testing whether the median of thepopulation distribution is different from zero.● Pr>=|S| is the p-value for the two-tailed signed rank test.● W:Normal = A statistic for testing whether the data follow a normal distribution.(Depending on the sample size, SAS uses different tests for normality, see manual.)This was requested by typing NORMAL in the PROC UNIVARIATE statement.● Prthe hypothesis of normality, while a high p-value does not necessarily indicate that thedata are normally distributed. Note that the tests for normality given by W dependhighly on the sample size!
การแปล กรุณารอสักครู่..
