Autocorrelation - Autocovariance - Auto/Serial Correlation/Covariance Function
Autocorrelation | the autocorrelation of a real or complex random process is the Pearson Correlation between values of the process at different times, as a function of the 2 times or of the time lag |
|---|---|
| Autocovariance Serial Covariance | the autocorrelation of a real or complex random process is the Covariation - Covariance between values of the process at different times, as a function of the 2 times or of the time lag |
NOTE: autocovariance is not well-defined for all-time series or processes, because the mean may not exist, or the variance may be zero (for a constant process) or infinite (for processes with distribution lacking well-behaved moments, such as certain types of power-law)
Auto-Correlation/Covariance - Functions/Formulas
Auto-Correlation/Covariance - Analysis/Tests
False Auto-Correlation (Auto-Correlation not caused by Population Auto-Correlation)
finding a non-zero Auto-Correlation does not mean it is true population Auto-Correlation, it could mean your model is missing an important input variable. A variable that correlates with one of the model independent variables
Dealing with Serial Correlation in Least Squares Estimators
serial correlation breaks one of the Gauss-Markov Assumptions - OLS Assumptions and also causes OLS to no longer become the best linear estimator BLUE
, multiple selections available,