Time series, normal distribution?

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
2 replies · 3K views
MaxManus
Messages
268
Reaction score
1

Homework Statement



If I have a time series model

x_1 = my + epsilon_1
x_i = my + a(x_{i-1} - my) + epsilon_i

epsilon_i are iid standard normal.

Can I then say that
y = a_1 x_1+a_2 x_2 + ... a_n x_n

is multi normal?



The Attempt at a Solution



All the x-es are normal since they are built up of previous epsilons thich are independent and normal.

But the x-es are not independent so can I then say that a linear combination of them is normal
 
Physics news on Phys.org
absolutely, any linear combinations of joint Gaussian plus whatever constant is still joint Gaussian (in your case, iid Gaussian is a special joint Gaussian)
 
Thanks for the help.