The augmented Dickey-Fuller test is a statistical test used to assess whether a time series has a unit root and is therefore nonstationary.
$$$$
The basic regression
A common ADF specification is:
$$
\Delta y_t = \alpha + \beta t + \gamma y_{t-1} + \sum_{i=1}^{p}\delta_i \Delta y_{t-i} + \varepsilon_t
$$
The extra lagged differences “augment” the original Dickey-Fuller test by helping control serial correlation in the residuals.
What the test is asking
The key question is whether (\gamma = 0). If that null cannot be rejected, the series may contain a unit root. If the null is rejected, the series is treated as stationary around the relevant deterministic terms.
Why economists care
Many macroeconomic and financial time series are persistent. If researchers ignore nonstationarity, they can estimate misleading relationships. The ADF test is therefore part of the standard workflow for deciding whether to difference a series, model cointegration, or proceed with stationary methods.
Knowledge Check
### What is the ADF test mainly used for?
- [x] Testing whether a time series has a unit root
- [ ] Measuring price elasticity
- [ ] Auditing financial statements
- [ ] Estimating fiscal multipliers directly
> **Explanation:** The ADF test is a standard tool for assessing time-series stationarity.
### Why is the test called "augmented"?
- [x] Because it adds lagged differences to address serial correlation
- [ ] Because it always uses more than one variable
- [ ] Because it includes a budget constraint
- [ ] Because it measures higher inflation
> **Explanation:** The augmentation improves the underlying Dickey-Fuller regression by allowing richer short-run dynamics.
### Why does unit-root testing matter in econometrics?
- [x] Because nonstationary series can produce misleading regression results if handled incorrectly
- [ ] Because stationary data cannot be analyzed
- [ ] Because all macro series are identical
- [ ] Because differencing is always unnecessary
> **Explanation:** Time-series properties affect inference, model choice, and whether long-run relationships are meaningful.