AUTOCORRELATED DATA
CALCULATIONS ON VARIANCES: SOME BASICS Let X and Y be random variables COV=0 if X and Y are independent.
WHAT IF COV(X i, X i+1 ) > 0? 1.We calculate an AVG by adding X’s 2.The VAR of the AVG is bigger by COV(X i, X i+1 ) 3.The formula for VAR assumes COV(X i, X i+1 ) =0 4.The formula underestimates VAR of the AVG 5.The formula for the width of the CI gives too small a width 6.The CI does not cover the true with the advertized probability 7.Our conclusion has oversold accuracy
AUTOCORRELATED DATA Consider the formula, called the Auto- Regressive (Lag 1) Process
NORMAL(0, 1) INDEPENDENT
c=0.2
C=0.5
C=0.7
C=0.9
C=0.9, 200 sample
C=0.99
c=0.5
c=0.7
c=0.9
c=.99
The Test for Rank 1 Autocorrelation Ho: (1) = 0 Ha: (1) <> 0
STATISTICALLY SIGNIFICANT AUTOCORRELATION Lag 1 autocorrelation (1) estimated by r(1) Normal Mean Variance
So the quantity z below is N(0, 1), and can be compared to critical values, and p-values can be computed… Simplifies when we are testing (1) = 0 Remember that this is a classical “wrong-way” hypothesis test
Sample Results crho(1)zp-value