Forecasting with ARMA Models

  • Sanjiv Jaggia


Increasing awareness and data availability has made statistical analysis an integral part of managerial decision making. Forecasting, in particular, has become an essential skill for all financial analysts, economists, marketing and accounting professionals. Arguably all business students should be educated in forecasting techniques with an emphasis on ARMA modeling. While statistical packages such as Eviews have increasingly become user friendly, the discussion and the im-plementation of the important diagnostic analysis of ARMA models in popular undergraduate textbooks has not been to-tally satisfactory. This paper succinctly outlines this important diagnostic step for ARMA modeling with Eviews, using monthly data on 3-month government Treasury bills rate for illustration. The presentation is accessible to readers with an intermediate level of statistics.