CHAPTER 1: An introduction to time series and forecasting
CHAPTER 1: An introduction to time series and forecasting
CHAPTER 1: An introduction to time series and forecasting
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
The variance of ˆµ is<br />
we have V ar(ˆµ) → 0 as n → ∞.<br />
V ar(ˆµ) = 1 n 2 (V ar(X 1) + V ar(X 2 ) + ...V ar(X n )) = σ2<br />
n<br />
Imagine the situation where all the X i ’s are “perfectly” correlated, i.e.<br />
Cov(X i , X j ) = σ 2<br />
Corr(X i , X j ) = 1<br />
We still estimate µ by<br />
ˆµ = (X 1 + X 2 + · · · + X n )/n<br />
the variance is then<br />
V ar(ˆµ) = 1 n 2 V ar(X 1 + X 2 + · · · + X n )<br />
= 1 n∑<br />
n { V ar(X 2 i ) + 2 ∑ Cov(X i , X j )}<br />
i=1<br />
i