Palgrave Handbook of Econometrics: Applied Econometrics

(Grace) #1

912 Macroeconometric Modeling for Policy


Acknowledgments


We thank Q. Farooq Akram and Kerry Patterson for helpful comments and discussions.


Notes



  1. See, e.g., Hendry (1995a, Ch. 2.3 and 15.3) for a concise definition of structure as the
    invariant set of attributes of the economic mechanism.

  2. This line of thought may lead to the following practical argument against large-scale
    empirical models: since modeling resources are limited, and some sectors and activities
    are more difficult to model than others, certain euations of any given model are bound
    to have less structural content than others, i.e., the model as a whole is no better than
    its weakest (least structural) equation.

  3. See Nymoen (2005) for an analysis of a recent failure in inflation forecasting.

  4. Sub-sections 17.2.1–4 draw on Bårdsen, Hurn and Lindsay (2004).

  5. Presently, we let the unemployment rate be constant and disregard it for simplicity. We
    return to the role of the rate of unemployment in section 17.2.6.

  6. The distinction between the inferential and model validation facets of modeling is due
    to Spanos (2008), who conclusively dispels the charge that misspecification testing rep-
    resents an illegitimate “reuse” of the data already used to estimate the parameters of the
    statistical model; see also Hendry (1995b, pp. 313–14).

  7. It might be noted that the income tax rateT2 is omitted from the analysis. This sim-
    plification is in accordance with previous studies of aggregate wage formation (see, e.g.,
    Nymoen, 1990, and Nymoen and Rødseth, 2003, where no convincing evidence of
    important effects from the average income tax rateT2 on wage growth could be found).

  8. Note that, due to the log-form,ζ=is/( 1 −is), whereisis the import share in private
    consumption.

  9. Strictly, we take the expectation through in both equations.

  10. NAM is a model project which extends from the early econometric assessment of wage
    and price inflation in Nymoen (1991), further developed in Bårdsen, Fisher and Nymoen
    (1998), Bårdsen and Fisher (1999), and the monetary transmission model of Bårdsen and
    Klovland (2000). Earlier versions of the model are documented in Bårdsen and Nymoen
    (2001), Bårdsen, Jansen and Nymoen (2003) and Bårdsenet al.(2005).
    NAM is used for both research purposes and teaching. The macroeconomic data is
    from the model databases of Statistics Norway (KVARTS model) and Norges Bank (FPAS
    database). Specific versions of the model are currently operative for (a) econometric
    forecasts of the Norwegian macroeconomy (NAM-EF) and (b) model-based analysis of
    financial stability in Norway (NAM-FS).

  11. See http://folk.uio.no/rnymoen/normetrics_index.html.

  12. In practice, the policy instrument is the sight deposit rate set by the central bank, but
    since the sight deposit rate represents (banks’) marginal funding cost, changes in the
    sight rate are transmitted to the money market rate immediately.

  13. The size of the depreciation will depend upon the risk premium, and whether expecta-
    tions counteract or strengthen the initial effect of the interest rate cut, etc. (cf. section
    17.3.3).

  14. Although the derivations are presented for a single equation with exogenous regressors,
    for ease of exposition, the techniques are, of course, the same for systems.

  15. A full listing of variables is given in the appendix to this chapter.

  16. See, e.g., Ericsson (2001) for an accessible discussion of forecast uncertainty, and its
    presentation in published forecasts.

  17. Automatized econometric inflation forecasts have been published twice a year, starting
    in July 2004. The forecasts are automatized, with a minimum of intervention after the
    econometric specification of the forecasting mechanism is completed.

Free download pdf