Module 2.7 Serial Correlation Flashcards
serial correlation is also known as
autocorrelation
refers to the situation in which residual terms are correlated with one another
serial correlation
exists when positive regression error in one time period increases probability of observing positive regression error for next time period
positive serial correlation
exists when positive error in one period increases probability of observing negative error in next period
negative serial correlation
positive serial correlation typically has what effect on coefficient standard errors
too small
positive serial correlation results in what type of error
type 1 error
rejection of null when it is true
type 1 error
false positive
type 1 error
false negative
type 2 error
not rejecting the null when it is false
type 2 error
very common in economic and financial data
positive serial correlation
two methods to detect presence of serial correlation
- residual plots
2. Durbin Watson statistic
DW statistic
DW=∑t=2T(ε^t−ε^t−1)2∑t=1Tε^t2
if the sample size is very large, DW =
2(1-r) where r equals correlation coefficient between residuals from current and prior period
DW < 2 when?
if error terms are positively serially correlated
