I have implemented the dickey duller test without using libraries or tools to understand how exactly the working is. My dataset is airline dataset and have tried using the same in excel to plot and view the stationarity of the dataset. After log transformations and differencing followed, the data became stationary as expected. However my implementation for stationarity check using Dickey fuller fails to show the desired output. For alpha = 0.05 the critical values are -1.95 for all of the N but what I could get was positive values only, hence showing it non-stationary (after applying on log and differenced data). I have gone through the steps again to identify that in order for the t-stat to be -ve, either the numeratror or denomntr must be -ve which can never be the case in

enter image description here

enter image description here

How then can I get a negative value so that on comparison with table will reject my null hypothesis? I have tried linear regression OLS for obtaining theta or p_hat value.

Any advice on this will be greatly appreciated...


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.