1/4
Looks like no tags are added yet.
Name | Mastery | Learn | Test | Matching | Spaced | Call with Kai |
|---|
No analytics yet
Send a link to your students to track their progress
Wat is Time series ?
Time series data are observations collected from the same unit repeatedly over time (e.g., daily, monthly, yearly).
analyze patterns and dynamics over time
forecast future values.
Wat betekent de notatie {Y1, Y2,…,YT} in time series?
Dit betekent dat we een variabele Y observeren over meerdere tijdsperiodes.
(Y_t) = de waarde van variabele Y op tijdstip t
(t = 1,2,\dots,T) = de tijd loopt van periode 1 tot en met periode T
T = totaal aantal observaties in de tijdreeks
Kort gezegd: dit beschrijft één volledige tijdreeks van lengte T.
What makes time series data different from cross-sectional data?
1⃣ Observations are ordered in time
We observe one sample path:
{Y1, Y2,…,YT}
2⃣ Observations are dependent over time
Past values influence future values:
Corr {Y1, Y-1} ≠ 0
👉 This means the classical OLS independence assumption is violated:
𝐸 𝜀𝑡𝜀𝑠 = 0 for 𝑡 ≠ s
Why is Ordinary Least Squares (OLS) not sufficient for time series data?
Dit model moet je leren
ACF | PACF | Model |
|---|---|---|
decay | cutoff | AR |
cutoff | decay | MA |
decay | decay | ARMA |