A Consistent Nonparametric Test for Causality in Quantile



Thm/2 [ J2(Qθ) - J2(Qθ - Ct)]

= Op(CTh-m/2)

=op(1) .

(A.25)


[3] Thm/2 [J3(Qθ)- J3(Qθ -Ct)] = Op(1):

Noting that H3t (5, t, Qθ ) = 0 because of F(Qθ (xj )| Zj ) - θ = 0 for j = t, s, we have

J3(Qθ)-J3(Qθ-CT)

1 ʃ. ʃ. 1    (z -z A

=--1--∑ ∑ ɪ K z-÷- I

T(T -1) ⅛ ⅛ hm  L h )

× {F(Qθ(xt)-Ct |z,)-θ}{F(Qθ(xs)-CtZs)-θ}

1TT1(zzI

= ^~∑ ∑ ∑ mK l^tv^ I CT S (Qθ ( Xt )) S (Qθ ( Xs ))

T(T- 1) t=1 sth Lh )

= CT1S(Q(Xt ))S(Q( X ))f Z(z, )                               (A.26)

T t=1

Thus, we have

EJ3( Qθ ) - J3( Qθ - ' )

1T

≤ ΛCT-EfZ(z■,)

T t=1

≤ ΛCT1 Ef (z-,) + ΛCT1 Ef (z,)-f,(z,)

Tt=1                       Tt=1

1T1T                           2

≤ ΛCT-Efz(Zt) + ΛCT-E{f (Zt)-fz(Zt)}
Tt=1Tt=1

= O(CT2 )                                                            (A.27)

Finally, we have

Th""-[J3(Qθ)- J3(Qθ -Ct)]

= Op(Thm/2CT2)

17



More intriguing information

1. AJAE Appendix: Willingness to Pay Versus Expected Consumption Value in Vickrey Auctions for New Experience Goods
2. The name is absent
3. Achieving the MDGs – A Note
4. The name is absent
5. The name is absent
6. The name is absent
7. Unemployment in an Interdependent World
8. Discourse Patterns in First Language Use at Hcme and Second Language Learning at School: an Ethnographic Approach
9. Feature type effects in semantic memory: An event related potentials study
10. Survey of Literature on Covered and Uncovered Interest Parities