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. Mergers and the changing landscape of commercial banking (Part II)
3. The name is absent
4. The name is absent
5. ‘Goodwill is not enough’
6. The name is absent
7. ENERGY-RELATED INPUT DEMAND BY CROP PRODUCERS
8. The Social Context as a Determinant of Teacher Motivational Strategies in Physical Education
9. Sectoral specialisation in the EU a macroeconomic perspective
10. The ultimate determinants of central bank independence