Tutorial learning objectives - Week 8

By the end of this tutorial, you should be able to:

1. Explain the Concept of Stationarity* (20 mins)

• Define stationarity as a condition where the statistical properties of a time series do not depend on time.

• Identify whether a series is stationary using visual inspection of time plots.

2. Apply Transformations and Differencing to Achieve Stationarity

• Use log or power transformations to stabilize variance in a time series.

• Apply first-order or seasonal differencing to remove trends and seasonality.

• Explain the purpose and effect of these techniques on the series.

3. Express Time Series Operations Using Backshift Notation

• Write the series using lag notation and backshift operator (B).

4. Use the ACF and PACF to Diagnose Time Series Properties

• Analyze the Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) of a series.

• Distinguish between stationary series, white noise, and potential AR/MA model structures using these tools.