Palgrave Handbook of Econometrics: Applied Econometrics

Gunnar Bårdsen and Ragnar Nymoen 907

–20

–10

2003

20

10

0

2004 2005 2006 2007

–5

–0

5

2003

20

15

10

2004 2005 2006 2007

–8

–4

2003

12

4

8

0

2004 2005 2006 2007

–20

–10

2003

30

20

0

2004 2005 2006 2007

10

1

4

3

2

2003

7

6

5

2004 2005 2006 2007

–20

–10

2003

30

20

0

2004 2005 2006 2007

10

60

70

2003

110

100

80

2004 2005 2006 2007

90

–20

–10

2003

30

20

0

2004 2005 2006 2007

10

–5

–10

5

0

2003

15

10

2004 2005 2006 2007

(a) Consumer price inflation (b) Unemployment rate (c) Wage growth

(d) GDP growth

(g) Money market interest rate (h) Real credit growth (i) Currency depreciation rate

(e) Import price inflation (f) Real exchange rate

Figure 17.14 Dynamic dEqCM forecasts 2003(1)–2007(3), with end-of-sample for estimation
of parameters in 2002(4)
Actual values are shown by solid lines, and forecasts by dashed lines. The distance between the two dotted
lines represents 70% prediction intervals.

Panels (a) and (b) show the NAM forecasts. The forecasts for 2007(1) are the same
as in Figure 17.16 for these two variables, but 2007(2) and 2007(3) are different since
the NAM forecasts are now conditional on first- and second-quarter information
for exogenous and predetermined variables (the coefficients are not updated). The
lack of adaptation to the location shifts of the growth rate of the actual series is
apparent. Panels (c) and (d) show the corresponding partial dEqCM forecasts. For
GDP growth there is less underprediction already in 2007(1), and in 2007(2) the
forecast has adapted. Panel (c) shows a marked improvement in adaptation also in
the credit growth rate, although not before the second quarter of 2007(2).

17.5 Conclusion

This chapter has presented a case for continuing to use macroeconometric models
for policy analysis, including such analyses that rely on instrument use to attain a