Ljung-Box test for Autocorrelation

DoubleDip Wrote:
——————————————————-
> yes, please see this:
> http://en.wikipedia.org/wiki/Portmanteau_test
>
> it’s a very easy test statistic to compute. here
> is the original paper
>
> http://www-stat.wharton.upenn.edu/~steele/Courses/
> 956/ResourceDetails/TestingNormality/LjungBox.pdf
>
> want a simple example using excel? see this
>
> http://books.google.com/books?id=gcspLg1fR7EC&pg=P
> A82&lpg=PA82&dq=ljung+box+excel&source=bl&ots=VoLn
> 2kwr9R&sig=apfiGdv4zSjLfCv4anUeHpx6C9o&hl=en&ei=m_
> 7hSqe1A8velAfw5NCKBw&sa=X&oi=book_result&ct=result
> &resnum=3&ved=0CA8Q6AEwAg#v=onepage&q=&f=false
Yes my own google search turned up all of those links along with many more that I read before posting this question to the forum. The problem is…and I hate to admit this…..I slept through statistics in college and as that troll pixel has pointed out on many occassions, CFA Statistics is pretty much a joke. I am using the test statistic to determine if my return series demonstrates autocorrelation and whether I should be unsmoothing the series before calculating things like standard deviation but I am not confident in my ability to interpret things correctly. My hope was to find someone that can A: tell me if I am interpreting it correctly, and B: If I am using it correctly.
Thanks for the links though.
 
Back
Top