Lecture 2

• Autocorrelation


  Degree of similarity between a given time series and a lagged version of itself over successive time intervals. It measures the relationship between a variable's current value and its past values.

  
• Autocorrelation


  • +1 = perfect positive correlation
  • -1 = perfect negative correlation
• Autocorrelation is a common pitfall with time series data
• Strict exogeneity: there is no systematic relationship between ut and X: E(ut|X) = 0 for t = 1, 2, ..., T
• No autocorrelation in the error term: cov(ut, ut-j|X) = 0, t ≠ j
• Strict exogeneity


  There is no systematic relationship between ut and X, Z, ...: E(ut|X, Z, ...) = 0 for t = 1, 2, . . . , T
• u must be uncorrelated with all Xs, all Zs... (past, current and future)
• b1 is biased towards zero if its true value is one when Yt has a stochastic trend