How to determine what the Seemingly Unrelated Regression error means in R
Posted
by
user2154571
on Stack Overflow
See other posts from Stack Overflow
or by user2154571
Published on 2013-07-02T23:00:49Z
Indexed on
2013/07/02
23:05 UTC
Read the original article
Hit count: 382
Filed under:
r
I'm using the systemfit() function to conduct a seemingly unrelated regression and am getting the following error:
Error in solve(sigma, tol = solvetol) : Lapack routine dsptrf returned error code 1
Yet, I'm unable to find the meaningful interpretation of what the error suggests is going on. Below is some simulated code that works to show what functions I'm using (the simulated code does not produce an error). Thanks for thoughts on this error.
y <- sample(seq(1:4), 100, replace = TRUE)
x1 <- sample(seq(0:1), 100, replace = TRUE) -1
x2 <- sample(seq(0:1), 100, replace = TRUE) - 1
x3 <- sample(seq(1:4), 100, replace = TRUE)
frame <- as.data.frame(cbind(y,x1,x2, x3))
mod_1 <- y ~ x1 + x3 + x1:x3
mod_2 <- y ~ x2 + x3 + x2:x3
output <- systemfit(list(mod_1, mod_2), data = frame, method = "SUR")
© Stack Overflow or respective owner
