Skip to main content

5.8 Statistical Tests

5.8.1
 
Standard post-estimation tests should be used to check that the underlying assumptions are appropriate for all types of macro models. The set of appropriate tests should be based on best practices in the relevant field / literature. The model development documentation should clearly indicate (i) the chosen test for each property, (ii) the nature of the H0 hypothesis, (iii) the cut-off values chosen upfront to determine the rejection or non-rejection.
 
5.8.2
 
In the context of time series regression, regression coefficients should be significant and residuals should be tested for autocorrelation and normality. The table below indicates properties that should be tested, at a minimum. Other tests may be considered, if necessary.
 

 
Table 9: Typical statistical tests for models based on time series regression
 
Property to be testedDescription of the property to be rejectedSuggested test (others may exist)
StationarityAbsence of stationarity in each time seriesAugmented Dickey-Fuller (ADF)
Co-integrationAbsence of stationarity in a linear combination of the dependent variable and each independent variableEngle-granger two-step method
MulticolinearityHigh correlation between the independent variablesVariance Inflation Factor
Coefficient significanceThe coefficients are not statistically significantly different from zeroCoefficient p-value on a t-distribution
AutocorrelationHigh correlation between the error terms of the modelLjung-Box test
HeteroscedasticityAbsence of relationship between independent variables and residualsBreusch-Pagan or White test
NormalityNormal distribution of the residualsShapiro Wilk