Error in fitting Natural Spline in R [closed]

Question

Closed. This question needs details or clarity. It is not currently accepting answers.

---

Want to improve this question? Add details and clarify the problem by editing this post.

Closed 8 years ago.

Improve this question

I'm fitting a natural spline model and I get the errors below. Can you help me understand and solve the error below?

Error in double(nx * ncoef) : vector size cannot be NA

In addition: Warning message: In nx * ncoef : NAs produced by integer overflow

Unfortunately, I cannot upload the dataset and unfortunately, you could not replicate my code. Instead, I'll try to explain my code as much as I can:

Below is the code that I run to fit a natural spline fit:

fit.temp <- lm(y ~ ns(x,knots = seq(1, nrow(data), by = 10)),data = data)

In my dataset, x is a sequence from 1 to the number of rows in data which is 424742.

I know this question is somewhat a vague question, but after spending a lot of energy, I didn't have any luck figuring out what this error is and how to solve it.

Thanks very much,

Answer 1

Somewhere within the code, it is trying to multiply two integers but the result is greater than the maximum integer that can be stored on your machine (.Machine$integer.max).

Without delving too much into the code, your inputs must be too big for this tool. Either work with a smaller data, or fewer knots by increasing the by argument to seq. Roughly, (nrow(data)^2 / .Machine$integer.max should give you a lower bound on how big by should be.