Autoregressive processes and stationarity • Stationarity is a desirable property of an estimated AR model for several reasons.
• One important reason is that a non-stationary AR
process exhibits the property that previous values of a time series will have a non-declining effect on the current value of the series as time progresses.
• In contrast, the autocorrelations of a stationary AR
process will decline eventually as the lag length is increased (the autocorrelation function (ACF) will decay geometrically to zero). Characteristic equation Integrated processes and differencing Unit root tests Illustration Augmented Dickey-Fuller (ADF) test in EViews Double click on sesame: You will see a spreadsheet containing the price series. Click on View and then select Unit Root Test… In the pop-up window (titled ‘Unit Root Test’), select Trend and intercept. Leave the other options as they are. Since the series is non-stationary (has unit root), we need to check if first differencing induces a stationary series. Click on View, select Unit Root Test… and then 1st difference. Non-stationarity and spurious regression • There are several reasons why the concept of stationarity is important.
• One of reasons is that the use of non-stationary data can lead
to spurious regressions.
• If two stationary variables are generated as independent
random series and when one of those variables is regressed on the other:
– the t-ratio on the slope coefficient would be expected not
to be significantly different from zero,
– and the value of R-squared would be expected to be very
low. • However, if two variables are trending over time, a regression of one on the other could have a high R-squared even if the two are totally unrelated.
• So, if standard regression techniques are applied to non-
stationary data, the end result could be:
– a regression that ‘looks’ good under standard measures
(significant coefficient estimates and a high value of R-squared),
– but which is really valueless or has no meaningful economic
interpretation.
• Such a model would be termed a ‘spurious regression’.
Illustration • We can see that R-squared is large (73%) and the model is adequate as judged by the F-test (p-value < 0.001).
• Moreover, the t-test indicates that the total reserves of
Ethiopia is significant at the 1% level, that is, total reserves of Ethiopia has a significant influence on US GDP.
• But we know that the total reserves of Ethiopia (a small
economy) can never affect the US economy (as measured by US GDP).
• This is clearly a spurious regression.
• Since both series are non-stationary (have unit roots), we need to take first differences.
• Then we have to check if first differencing removes the
unit root problem.
• If this is so, US GDP and the total reserves of Ethiopia
are integrated of order one (I(1)). • We can see that: • R-squared is zero • The F-statistic is insignificant (p-value > 0.05). • The t-ratio is insignificant (p-value > 0.05).
• Thus, there is no relationship between US GDP and
the total reserves of Ethiopia.
• The significant relationship that we obtained earlier was
simply a consequence of the underlying trend in both series.