ETC2410 Lecture Notes - Lecture 7: Dependent And Independent Variables, Time Complexity, Seasonal Adjustment
13 Nov 2018
School
Department
Course
Professor
Document Summary
It only makes it hard to ascertain the contribution of each one x in a group of (cid:373)ulti(cid:272)olli(cid:374)ear (cid:454)(cid:859)s to e(cid:454)plai(cid:374)i(cid:374)g (cid:455) If a constant and all 4 dummy variables are used, we fall into the dummy variable trap => (cid:454)(cid:859)s are perfe(cid:272)tl(cid:455) (cid:272)olli(cid:374)ear: nb: a (cid:373)ore a(cid:272)(cid:272)urate esti(cid:373)ate of log(cid:894)(cid:455)(cid:895) give(cid:374) (cid:454) = 1. Application of dummy variables to time series: seasonality, many business and economic time series exhibit a trend and regular cyclical behaviour that is repeated with regular periodicity i. e. seasonality. Important not to mistake seasonal movements as movements in trend or business cycle. Seasonal adjustment - removing seasonality from the data: however, in some applications (e. g. forecasting), seasonality is important and must be handled accordingly. Australian economy: define a dummy variable, postgfc, which equals 1 on and after the 3rd quarter of 2008 and 0 before that, run a regression of the australian growth rate on a constant and postgfc.
