ECON
TimeSeriesBook.pdf

Suppose that x t follows an arma process φl x t c

Info icon This preview shows pages 167–169. Sign up to view the full content.

Suppose that { X t } follows an ARMA process Φ(L)∆ X t = c + Θ(L) Z t then μ t = δ +Ψ(1) Z t can be identified as the trend component. This means that the trend component can be recursively determined from the observations by applying the formula μ t = Φ(L) Θ(L) Ψ(1) X t . The cyclical component is then simply the residual: ε t = X t - μ t . In the above decomposition both the permanent (trend) component as well as the stationary (cyclical) component are driven by the same shock Z t . A more sophisticated model would, however, allow that the two compo- nents are driven by different shocks. This idea is exploited in the so-called structural time series analysis where the different components (trend, cycle,
Image of page 167

Info icon This preview has intentionally blurred sections. Sign up to view the full version.

7.2. OLS ESTIMATOR WITH INTEGRATED VARIABLES 149 season, and irregular) are modeled as being driven by separated shocks. As only the series { X t } is observed, not its components, this approach leads to serious identification problems. See the discussion in Harvey (1989), Han- nan and Deistler (1988), or Mills (2003). In Sections 17.1 and 17.4.2 we will provide an overall framework to deal with these issues. Examples Let { X t } be a MA(q) process with ∆ X t = δ + Z t + . . . + θ q Z t - q then the persistence is given simply by the sum of the MA-coefficients: Ψ(1) = 1+ θ 1 + . . . + θ q . Depending on the value of these coefficients. The persistence can be smaller or greater than one. If { X t } is an AR(1) process with ∆ X t = δ + φ X t - 1 + Z t and assuming | φ | < 1 then we get: ∆ X t = δ 1 - φ + j =0 φ j Z t - j . The persistence is then given as Ψ(1) = j =0 φ j = 1 1 - φ . For positive values of φ , the persistence is greater than one. Thus, a shock of one is amplified to have an effect larger than one in the long-run. If { X t } is assumed to be an ARMA(1,1) process with ∆ X t = δ + φ X t - 1 + Z t + θZ t - 1 and | φ | < 1 then ∆ X t = δ 1 - φ + Z t +( φ + θ ) j =0 φ j Z t - j - 1 . The persistence is therefore given by Ψ(1) = 1 + ( φ + θ ) j =0 φ j = 1+ θ 1 - φ . The computation of the persistence for the model estimated for Swiss GDP in Section 5.6 is more complicated because a fourth order difference 1 - L 4 has been used instead of a first order one. As 1 - L 4 = (1 - L)(1+L+ L 2 + L 3 ), it is possible to extend the above computations also to this case. For this purpose we compute the persistence for (1 + L + L 2 + L 3 ) ln BIP t in the usual way. The long-run effect on ln BIP t is therefore given by Ψ(1) / 4 because (1 + L + L 2 + L 3 ) ln BIP t is nothing but four times the moving- average of the last four values. For the AR(2) model we get a persistence of 1.42 whereas for the ARMA(1,3) model the persistence is 1.34. Both values are definitely above one so that the permanent effect of a one-percent shock to Swiss GDP is amplified to be larger than one in the long-run. Campbell and Mankiw (1987) and Cochrane (1988) report similar values for the US.
Image of page 168
Image of page 169
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern