martin.g25
New member
- Joined
- Feb 10, 2023
- Messages
- 6
Hello,
I would like to ask for a help in relation to the Augmented Dickey Fuller test interpretation. I have to perform ADF tests for a group of macroeconomic variables and I would have two questions:
1. how many lags should be included in the ADF test? I read that in case of quarterly data, we should include 4 lags. In case of monthly data we should include 12 lags. Is it true or do you know any other rule?
2. which lag should I consider in my final test decision? For example, I test GDP and for lags 0,1 and 3 I would reject the null hypotesis and I could say that the time series is stacionary. However, for lags 2 and 4 I fail to reject the null hyposes and I must say that the GPD is non-stationary? What is true?
Please see attached image to see what I mean.
Could you please help me in this problem?
Thank you
I would like to ask for a help in relation to the Augmented Dickey Fuller test interpretation. I have to perform ADF tests for a group of macroeconomic variables and I would have two questions:
1. how many lags should be included in the ADF test? I read that in case of quarterly data, we should include 4 lags. In case of monthly data we should include 12 lags. Is it true or do you know any other rule?
2. which lag should I consider in my final test decision? For example, I test GDP and for lags 0,1 and 3 I would reject the null hypotesis and I could say that the time series is stacionary. However, for lags 2 and 4 I fail to reject the null hyposes and I must say that the GPD is non-stationary? What is true?
Please see attached image to see what I mean. Could you please help me in this problem?
Thank you