QE Time Series Flashcards
Cointegration
when two or more time series variables share a common stochastic trend
Common trend
A trend shared by two or more time series
Dickey Fuller Test
A method for testing for a unit root in a first order regression AR(1) model
Distributed lag model
a regression model in which the regressors are current and lagged values of X
Endogenous variable
A variable that is correlated with the error term
Error Term
The difference between Y and the population regression function
autocorrelation
the correlation between a time series variable and its lagged value
Autocovariance
The covariance between a time-series variable to its past (that is lagged) values
correlation
the unit free measure of the extent to whch two variables move together
Forecast Interval
An interval that contains the future value of a time series variable with a prescribed probability
Granger Causality Test
A procedure for testing whether current and lagged values of one time series help predict future values of another time series
Homoskedasticity
The variance of the error term, conditional on regressors, is constant
Heteroskedacity
The variance of the error term conditional on regressors is not constant
imperfect multicolliniarity
the condition in which two or more regressors are highly correlated
Instrumental variables regression
A way to obtain a consistent estimator of the unknown coefficients of the population regression function when the regressor, X, is correlated with the error term
Non-stationary
the joint distribution of time series variable and it’s lags changes over time
OLS residual
The difference between Y and the OLS regression line
Omitted Variable Bias
The bias in an estimator that has arisen because a variable that is a determinant of y and is correlated with the regressor has been omitted from the regression
Order of integration
The number of times a time series variable must be differenced to make it stationary. A time series variable that is integrated of order p must be differenced p times and is denoted I(p)
p value
the smallest significance level at which the null hypothesis can be rejected.
perfect multicollinearity
One of the regressors is the exact linear function of another of the regressors
Pseudo out of sample forecasting
A forecast computed over part of a sample using a procedure as if the sample has not yet been realised
R^2
In a regression, the fraction of the sample variance of the dependent variable that is explained by the regressor
Random Walk
A time series regression where the value of the variable equals the value of it in a previous period plus a random error
Random walk with drift
a random walk in which the change in variable has a non-zero mean but it otherwise unpredictable
Randomized controlled experiment
participants are randomly assigned to a control group which receives no treatment, or a treatment group, which receives a treatment.
serially uncorrelated
a time series variable where all autocorrelations equal zero
simultaneous causality bias
When, in addition to the causal link of interest from X to Y, there is a causal link from Y to X. Simultaneous causality makes X correlated with the error term in the population regression of interest
Stationary
The joint distribution of a time series variable and it’s lagged values does not change over time
Stochastic trend
A persistent but random long-term movement of a variable over time
Type 1 error
The null hypothsis is TRUE but it’s REJECTED
type 2 error
The null hypothesis isn’t rejected but it’s FALSE
Granger causality test
In an ADL model, an indication of how useful X values are in predicting Y. It’s an F-test
ADL Model
Autoregressive Distributed Lag model, where there is another variable in an AR(p) model
Mean Squared Forecast Error
Expected value of the squared difference between the fitted values implied by the predictive function B(hat) and the unobserved function B