Scribd
Upload
104 views3 pages

Answer Key To Stock and Watson Chapter 16 Questions

R1t follows a random walk process and is integrated of order one, I(1). Rkt is calculated as the average of R1t over k periods plus an error term et. Since both Rkt and R1t are I(1) but their difference Rkt - R1t = et is I(0), they are cointegrated with a cointegrating coefficient of 1. The i-period ahead forecast of R1t depends on its current level and the forecast of the difference term ΔR1t, which follows an AR(0.5) process. As a result, the i-period ahead forecast of Rkt also depends on R1t and ΔR1t.

Uploaded by

pnunns
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
104 views3 pages

Answer Key To Stock and Watson Chapter 16 Questions

R1t follows a random walk process and is integrated of order one, I(1). Rkt is calculated as the average of R1t over k periods plus an error term et. Since both Rkt and R1t are I(1) but their difference Rkt - R1t = et is I(0), they are cointegrated with a cointegrating coefficient of 1. The i-period ahead forecast of R1t depends on its current level and the forecast of the difference term ΔR1t, which follows an AR(0.5) process. As a result, the i-period ahead forecast of Rkt also depends on R1t and ΔR1t.

Uploaded by

pnunns
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

. 16.

(a) BecauseR1tfollowsarandomwalk(R1tR1t1ut),theiperiodaheadforecastofR1tis
R1t i|t R1t i 1|t R1t i 2|t L L R1t .
Thus
1 k 1 k
Rkt
k i 1
R1t i|t et R1t et R1t et .
k i 1

(b) R1tfollowsarandomwalkandisI(1).RktisalsoI(1).GiventhatbothRktandR1tare
integratedoforderone,andRktR1tetisintegratedoforderzero,wecanconcludethat

RktandR1tarecointegrated.Thecointegratingcoefficientis1.
R1t 0.5 ut , R1t
(c) When isstationarybutR1tisnotstationary.
R1t 1.5R1t 1 0.5R1t 2 ut ,
anAR(2)processwithaunitautoregressiveroot.Thatis,
I (1) R1t
R1tis .Theiperiodaheadforecastof is
R1t i|t 0.5R1t i 1|t 0.52 R1t i 2|t L L 0.5i R1t .

TheiperiodaheadforecastofR1tis
R1t i t R1t i 1|t R1t i|t
R1t i 2|t R1t i 1|t R1t i|t
K K
R1t R1t 1|t L R1t i|t
R1t (0.5 L 0.5i ) R1t
0.5(1 0.5i )
R1t R1t .
1 0.5

Thus
1 k 1 k
Rkt
k i 1
R1t i t et [ R1t (1 0.5i )R1t ] et
k i 1
R1t R1t et .

1k ik1 (1 0.5i ). Rkt R1t R1t et .


where Thus ThusRktandR1tarecointegrated.
Thecointegratingcoefficientis1.
(d) WhenR1t0.5R1t1ut,R1tisstationaryanddoesnothaveastochastictrend.
R1t i|t 0.5i R1t , Rkt R1t et , 1k ik1 0.5i.
sothat, where SinceR1tandetareI(0),
thenRktisI(0).
16.6. (a) Rewritetheregressionas
Yt3.02.3Xt1.7(Xt+1Xt)0.2(XtXt1)ut
Thus2.3,11.7,00.2and10.0.
(b) CointegrationrequiresXttobeI(1)anduttobeI(0).
(i) No
(ii) No
(iii) Yes

You might also like