帮顶

这问题有意思

需要请联系QQ：1298997509

下面是协整的Johansen的LM test程序，看看没有么帮助：

johansen.test<-
function(x, L = 2)
{
        x <- ts(x)
        n <- nrow(x)
        p <- ncol(x)
        Ly <- lag(x[, 1], -1)
        D <- diff(x[, 1])
        for(i in 1:p) {
                if(i > 1) {
                        D <- ts.intersect(D, diff(x[, i]))
                        Ly <- ts.intersect(Ly, lag(x[, i], -1))
                }
                if(L > 0)
                        for(j in 1:L)
                                D <- ts.intersect(D, lag(diff(x[, i]),  - j))
        }
        iys <- 1 + (L + 1) * (0:(p - 1))
        Y <- D[, iys]
        X <- D[,  - iys]
        Ly <- ts.intersect(Ly, D)[, 1:p]
        ZD <- lm(Y ~ X)$resid
        ZL <- lm(Ly ~ X)$resid
        df <- nrow(X) - ncol(X) - 1
        S00 <- crossprod(ZD)/df
        S11 <- crossprod(ZL)/df
        S01 <- crossprod(ZD, ZL)/df
        M <- solve(S11) %*% t(S01) %*% solve(S00) %*% S01
        eigen(M)$values   #返回特征值
}

关于该检验的部分理论见http://en.wikipedia.org/wiki/Johansen_test

grangertest: grangertest is simply a Wald test comparing the unrestricted model—in which y is explained by the lags (up to order order) of y and x—and the restricted model—in which y is only explained by the lags of y.
causality:Two causality tests are implemented. The first is a F-type Granger-causality test and the second is a Wald-type test that is characterized by testing for nonzero correlation between the error processes of the cause and effect variables.