Nonparametric cointegration analysis



nξ ξ TMnz ( F) ξ ξ IF(1)

I


snw(1)
nn


(A.13)


.. „            /-f          c

ξ T(z0 - w 0)Vn F(1) - Jfx)

Note that

F(1) -jxf(x ) dx = ʃF(x ) dx = 0,
hence

F(1) - ʃInxlf(x)dx 1 fxfx)|dx
J n             n J

[nx ] dx 1
n


(A.14)


(A.15)


and consequently equation (A.13) then becomes

nξ ξ TMnz ( F) - ξ ïF(1)

I


Snw(1)
nn


- ʃfx) S^ dx 1
nn ^


(A.16)


Moreover,

ξTS(x) - ξ⅛∙m - ξTw0.                                                       (A'17)
and consequently

n ξτmδ(F) = ξ7(F(1)(wn~w0) - ʃʃ(x)(w[nx] ~w0)dx]

(A.18)


ξ ξ ^F(1)Wn - f/(.x )W [ nx ] dx) = F(1) ξ Twn + θp (1)-

The last equality in (A.18) follows from the fact that by the dominated convergence theorem,

40



More intriguing information

1. The name is absent
2. The name is absent
3. BEN CHOI & YANBING CHEN
4. The name is absent
5. Behavioural Characteristics and Financial Distress
6. The Determinants of Individual Trade Policy Preferences: International Survey Evidence
7. The name is absent
8. Yield curve analysis
9. The name is absent
10. Imitation in location choice
11. The name is absent
12. The name is absent
13. The name is absent
14. Washington Irving and the Knickerbocker Group
15. Labour Market Flexibility and Regional Unemployment Rate Dynamics: Spain (1980-1995)
16. Death as a Fateful Moment? The Reflexive Individual and Scottish Funeral Practices
17. Concerns for Equity and the Optimal Co-Payments for Publicly Provided Health Care
18. Optimal Tax Policy when Firms are Internationally Mobile
19. The name is absent
20. The urban sprawl dynamics: does a neural network understand the spatial logic better than a cellular automata?