1

I searched multiple entries with similar topic, but couldn't find suitable answer. I want to compare two time series, interest rate and investment grade corporate bond yield over 15 years. I have quarterly data for interest rate. And I want to also control for other variables like GDP, inflation and so on. Which statistical model will suite the best? My guess is ARIMA, but any input will be appreciated very much. Thank you.

Farhad M.
  • 25
  • 2

1 Answers1

1

Multivarite ARIMA ( single equation) is called a Transfer Function/Dynamic Regression which is probably what you are looking for. You need to determine appropriate contemporaneous and lag structure while incorporating possible Intervention Detected structure. like pulses /step/level shifts, local time trends and seasonal pulses. One place to get an intro is here ARIMAX model's exogenous components? and here How to forecast a time series which is dependent on different time series? and http://www.autobox.com/cms/index.php/afs-university/intro-to-forecasting/doc_download/24-regression-vs-box-jenkins and here https://onlinecourses.science.psu.edu/stat510/node/75/. If you wish to post your data in a csv or xls format I might be able to help you further.

Community
  • 1
IrishStat
  • 27,906
  • 5
  • 29
  • 55
  • Thank you very much for your feedback. I dont have complete dataset yet, but will post it here once I have it. – Farhad M. Jan 10 '17 at 19:38