Lecture 4 Key terms

Question 1

highly persistent time series

Answer 1

a time series in which the value of y today is important to determine the value of y far in the future

Question 2

unit root (process)

Answer 2

(an ar(1) process for which) the stability condition does not hold, ie p = 1

Question 3

(first) difference stationary / I(1) process

Answer 3

(first) differencing turns a non stationary unit root process into a weakly dependent process

Question 4

trend stationarity

Answer 4

removing the trend makes the process stationary

Question 5

memory

Answer 5

property that determines for how many periods a shcom affects the variable

Question 6

dickey fuller test

Answer 6

test that determines whether a process contains a unit root

Question 7

augmented df test

Answer 7

test used to determine if a process has a unit root if the process has serially correlated errors

Question 8

(low) power

Answer 8

(an increased) chance of accepting the null when the alternative is true

Question 9

structural break

Answer 9

the data is split on some way due to collection technique or some natural quality of the data

Question 10

spurious regression

Answer 10

an unobserved variable is correlated with both the explanatory variables and the error terms, leading us to infer a causal relationship between our exp variable and dep variable that may not exist