Nonparametric cointegration analysis



Note that Assumption 2 guarantees that the matrix Vr,m is a.s. nonsingular.
Denoting

Xk - RRTkc(1)C(1)TR J'4RqT,C(1)Xk, Yk = RTC(1)C(1)tXj,J'÷Rr(1) Y,,
and using the result of Andersen, Brons and Jensen (1983), it follows straightforwardly from

Lemma 4:

Theorem 1. Let λ1 ,m .. ≥ λq,m be the ordered solutions of the generalized eigenvalue problem

(19)


defA λ λ(B +n 2A 1)1 = 0,
|_ m mm          m 7J ’

and let λ1,m... ≥ λq-r,m be the ordered solution of the generalized eigenvalue problem

m

vk1


(20)

where the X*i ’s and Y*j ’s are i.i.d. Nq-r(0,Iq-r). Ifzt is cointegrated with r linear independent
cointegrating vectors then under Assumptions 1-3,
(λ1 ,m ,..,λq,m ) converges in distribution to
(λ1,m,...,λq-r,m,0,..,0).

In order to show how fast (λq-r+1 ,m ,..,λq,m ) converges to (0,....,0), observe from Lemma
4 that

n RR τA RR t(B +
m f mm


n2A l)RRtA R^'1
m 7 /
m f


R Rn-2a 1 r)'2Rt(B
           m f     × m


+ n -2a l)RRn -2a^1r) *
m 7 /            m f


D (O


O


Moreover, it is easy to see that the solutions μj,m of the generalized eigenvalue problem
i +n-2A-1)RRtaR'ɪ -μI

det -J-R tAR RR t(Bi


m             m 7 / n * T T

11



More intriguing information

1. The name is absent
2. The name is absent
3. Reputations, Market Structure, and the Choice of Quality Assurance Systems in the Food Industry
4. The name is absent
5. AN ANALYTICAL METHOD TO CALCULATE THE ERGODIC AND DIFFERENCE MATRICES OF THE DISCOUNTED MARKOV DECISION PROCESSES
6. Behaviour-based Knowledge Systems: An Epigenetic Path from Behaviour to Knowledge
7. Can genetic algorithms explain experimental anomalies? An application to common property resources
8. BODY LANGUAGE IS OF PARTICULAR IMPORTANCE IN LARGE GROUPS
9. New Evidence on the Puzzles. Results from Agnostic Identification on Monetary Policy and Exchange Rates.
10. New urban settlements in Belarus: some trends and changes