This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Ec 178 ECON & BUS FORECASTING Foster, UCSD HW #3 -- ANSWER SHEET How would you describe the non-stationarity? How could you transform y t into a stationary series? Do you recommend deterministic models here? Explain.  Non-constant mean (upward trend); possibly a random walk with positive drift; variance seems relatively constant. Try first differences or take residuals from linear trend curve. Deterministic curve fitting might be ok if there is some long-run equilibrium process associated with the variable. It is hard to tell. What is associated with this pattern in the ACF?  It dies out very slowly, so it indicates a random walk (unit root), a series that is not stationary. ______/ 75 50 100 150 200 Yt 20 40 60 80 100 t t = 1... 100 Figure 1 -- Raw Data-1.00-0.50 0.00 0.50 1.00 Autocorrelations of y 2 4 6 8 10 Lag Bartlett's formula for MA(q) 95% confidence bands Ec 178 HOMEWORK #3 KEY p. 2 Record your test statistic and the 5% critical value in Table A. What are the null and alternative hypotheses for this test? What is your conclusion...
View Full Document
This note was uploaded on 06/20/2008 for the course ECON 178 taught by Professor Foster during the Spring '08 term at UCSD.
- Spring '08