12.07.2015 Views

"Frontmatter". In: Analysis of Financial Time Series

"Frontmatter". In: Analysis of Financial Time Series

"Frontmatter". In: Analysis of Financial Time Series

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

34 LINEAR TIME SERIES ANALYSIS AND ITS APPLICATIONS(a)(c)acf-1.0 -0.5 0.0 0.5 1.00 5 10 15 20lagacf-1.0 -0.5 0.0 0.5 1.00 5 10 15 20lag(b)(d)acf-1.0 -0.5 0.0 0.5 1.00 5 10 15 20lagacf-1.0 -0.5 0.0 0.5 1.00 5 10 15 20lagFigure 2.4. The autocorrelation function <strong>of</strong> an AR(2) model: (a) φ 1 = 1.2 andφ 2 =−0.35,(b) φ 1 = 0.6 andφ 2 =−0.4, (c) φ 1 = 0.2 andφ 2 = 0.35, (d) φ 1 =−0.2 andφ 2 = 0.35.r t = 0.0047 + 0.35r t−1 + 0.18r t−2 − 0.14r t−3 + a t , ˆσ a = 0.0098. (2.12)Rewriting the model asr t − 0.35r t−1 − 0.18r t−2 + 0.14r t−3 = 0.0047 + a t ,we obtain a corresponding third-order difference equationwhich can be factored as(1 − 0.35B − 0.18B 2 + 0.14B 3 ) = 0,(1 + 0.52B)(1 − 0.87B + 0.27B 2 ) = 0.The first factor (1 + 0.52B) shows an exponentially decaying feature <strong>of</strong> the GNPgrowth rate. Focusing on the second-order factor 1 − 0.87B − (−0.27)B 2 = 0, wehave φ 2 1 + 4φ 2 = 0.87 2 + 4(−0.27) =−0.3231 < 0. Therefore, the second factor <strong>of</strong>the AR(3) model confirms the existence <strong>of</strong> stochastic business cycles in the quarterlygrowth rate <strong>of</strong> U.S. real GNP. This is reasonable as the U.S. economy went through

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!