Hellow,
I have applied linear ARDL approach for a time series multivariate regression model but the model is suffering serial correlation problem. How can I remove this problem ? Knowing the serial correlation problem , I have applied NARDL approach and the serial correlation automatically removed from the model. Does serial correlation be a problem, in this case, though my target model is NARDL ? I am using STATA 15.
First, I assume ARDL stands for Autoregressive distributed lag model. and NARDL stands for Nonlinear ARDL. Now, a first order autoregressive model is y(t)=a+b*y(t-1) + eps. It thus seems hard to remove serial correlation from your model and leave much behind to model. Could you perhaps describe your research in enough detail so that we could at least understand your question.? Best, David Booth PS graphs are a wonderful tool. BTW see this link:https://www.researchgate.net/post/How_to_remove_serial_correlation_for_ARDL_model and this one: https://www.google.com/search?client=firefox-b-1-d&ei=74SpX5DIHcSDtQborKWYCg&q=NARDL&oq=NARDL&gs_lcp=CgZwc3ktYWIQAzICCAAyAggAMgYIABAHEB4yAggAMgYIABAHEB4yBggAEAcQHjIGCAAQBxAeMgYIABAHEB4yBggAEAcQHjIGCAAQBxAeUKP5DFij-Qxg05UNaABwAXgAgAGWCIgBlgiSAQM3LTGYAQCgAQGqAQdnd3Mtd2l6wAEB&sclient=psy-ab&ved=0ahUKEwjQ-7zFhvbsAhXEQc0KHWhWCaMQ4dUDCAw&uact=5
- Badges
- Science topic
- Similar topics
- Statistics
- Time Series