Time-Series Basics Flashcards
What is time series data?
It consists of observations of a variable over time at regular intervals (e.g daily, monthly, yearly)
what are lagged variables?
A lag refers to a past value of the variable
Yt-1 means the value of Y in the previous period
How are lags used?
Lags allow past values of a variable to influence the present. First differences (Δ𝑌𝑡) are used to measure change.
ΔYt =Yt−Yt−1
these differences are used to help remove trends and detect stationary
What is a static model?
A static model assumes that Yt is immediately affected by Xt
Yt=β0+β1Xt+ut
E.g todays stock price depends only on todays interest rate
What is a dynamic model?
A dynamic model allows past values (Xt-1) to influence Yt
Yt=β0+β1Xt+β2Xt−1+ut
Example: Todays stock price depends on today and yesterdays interest rate
Why would you use lags?
Some effects take time to materialise, for examples interest rate changes affecting GDP
Helps in forecasting future values
What is a Finite Distributed Lag (FDL) model?
It is a model that allows multiple past values (lags) of Xt to influence Yt
What is the FDL formula example?
Yt=β0+β1 Xt+β2Xt−1+β3X t−2+u t
If government spending (Xt) affects GDP growth (Yt) the effect might last for several periods
β1 = immediate effects of spending on GDP
β2 = effects from last years spending
β3 = effects from 2 years ago
Why is the FDL formula useful?
It helps analyse how long an effect lasts
It is used in policy analysis (e.g does a interest rate cut affect growth over 1 year or 3 years?
