require(MTS)
rt=rnorm(200)
b<-archTest(rt, lag = 10)
This code results in the values
Q(m) of squared series(LM test):
Test statistic: 7.694531 p-value: 0.6586466
Rank-based Test:
Test statistic: 20.80503 p-value: 0.02249487
But when the structure of b is checked using typeof(b) and str(b), it gives NULL.
How is it possible? How can one knows the structure of b as it is needed to extract values from this variable.
If you look at the source code of archTest, you're question boils down to more or less this one: can I stop cat from returning NULL? To which the answer is no.
You can modify archTest however such that it returns the test statistic and p-value (see below). b is a named vector.
b <- my_archTest(rt, lag = 10)
str(b)
# Named num [1:2] 7.565 0.671
# - attr(*, "names")= chr [1:2] "Test statistic" "p-value"
b
#Test statistic p-value
# 7.5653283 0.6712114
b["Test statistic"]
#Test statistic
# 7.565328
I added three lines at the end of MTS::archTest.
my_archTest <- function (rt, lag = 10) {
at = rt
if (is.matrix(at))
at = at[, 1]
m1 = acf(at^2, lag.max = lag, plot = F)
acf = m1$acf[2:(lag + 1)]
nT = length(at)
c1 = c(1:lag)
deno = rep(nT, lag) - c1
Q = sum(acf^2/deno) * nT * (nT + 2)
pv1 = 1 - pchisq(Q, lag)
cat("Q(m) of squared series(LM test): ", "\n")
cat("Test statistic: ", Q, " p-value: ", pv1, "\n")
rk = rank(at^2)
m2 = acf(rk, lag.max = lag, plot = F)
acf = m2$acf[2:(lag + 1)]
mu = -(rep(nT, lag) - c(1:lag))/(nT * (nT - 1))
v1 = rep(5 * nT^4, lag) - (5 * c(1:lag) + 9) * nT^3 + 9 *
(c(1:lag) - 2) * nT^2 + 2 * c(1:lag) * (5 * c(1:lag) +
8) * nT + 16 * c(1:lag)^2
v1 = v1/(5 * (nT - 1)^2 * nT^2 * (nT + 1))
QR = sum((acf - mu)^2/v1)
pv2 = 1 - pchisq(QR, lag)
cat("Rank-based Test: ", "\n")
cat("Test statistic: ", QR, " p-value: ", pv2, "\n")
out <- c("Test statistic" = QR, # new lines added here
"p-value" = pv2)
return(out)
# or instead simply
# c("Test statistic" = QR,
# "p-value" = pv2)
}
As pointed out by @42- in the comments, the return call is not necessary.