Theere are two methods given for serial correlation test for time series data: one is auto correlation test and one is dickey fuller test.
can anyone tell me that when we can’t use auto correlation test and so that we should use dickey fuller
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
We have to use autocorrelation when the trend of the model is mean reverting and autocorrelation can only be used when it’s a covariance stationary.. inorder to check for covariance stationary we use DF test which check whether b1 is equal to 1 or not if b1=1 then we can’t use autocorrelation as it’s not covariance stationary