EViews 7 New Econometrics and Statistics: Diagnostics

EViews 7 features a number of additions and improvements its extensive set of basic diagnostics. Notably additions include greatly expanded options for single equation robust covariances, a variety of new single-equation post-estimation diagnostics, and specialized diagnostics for equations estimated using instrumental variables and GMM.

Coefficient Covariance Calculation

EViews 7 offers an expanded choice of options for computing standard errors for single equation regression estimates.

There is now an option to turn off the degrees-of-freedom adjustment to standard errors.

More importantly, an expanded range of HAC covariance options mirrors those for the stand-alone covariance calculations. You may compute symmetric or one-sided long-run covariances using nonparametric kernel (Newey-West 1987, Andrews 1991), parametric VARHAC (Den Haan and Levin 1997), and prewhitened kernel (Andrews and Monahan 1992) methods. In addition, EViews supports Andrews (1991) and Newey-West (1994) automatic bandwidth selection methods for kernel estimators, and information criteria based lag length selection methods for VARHAC and prewhitening estimation. The new options may be found by selecting HAC in the Coefficient covariance matrix combo box on the Options page of the Equation dialog, and then pressing the HAC Options button.

Expanded Post-Estimation Diagnostics

• The new Scaled Coefficients view displays the coefficient estimates, the standardized coefficient estimates and the elasticity at means. The standardized coefficients are the point estimates of the coefficients standardized by multiplying by the standard deviation of the dependent variable divided by the standard deviation of the regressor.
  The elasticity at means are the point estimates of the coefficients scaled by the mean of the dependent variable divided by the mean of the regressor.
• The Confidence Intervals view displays a table of confidence intervals for each of the coefficients in the equation.
  The Confidence Intervals dialog allows you to enter the size of the confidence levels. These can be entered a space delimited list of decimals, or as the name of a scalar or vector in the workfile containing confidence levels. You can also choose how you would like to display the confidence intervals. By default they will be shown in pairs where the low and high values for each confidence level are shown next to each other.
• EViews 7 now displays Variance Inflation Factors. Variance Inflation Factors (VIFs) are a method of measuring the level of collinearity between the regressors in an equation. VIFs show how much of the variance of a coefficient estimate of a regressor has been inflated due to collinearity with the other regressors.
• The new Coefficient Variance Decomposition view of an equation provides information on the eigenvector decomposition of the coefficient covariance matrix. This decomposition is a useful tool to help diagnose potential collinearity problems amongst the regressors. The decomposition calculations follow those given in Belsley, Kuh and Welsch (2004).
• Influence statistics are a method of discovering influential observations, or outliers. They are a measure of the difference that a single observation makes to the regression results, or how different an observation is from the other observations in an equation’s sample. EViews provides a selection of six different influence statistics: RStudent, DRResid, DFFITS, CovRatio, HatMatrix and DFBETAS.
• Leverage plots are the multivariate equivalent of a simple residual plot in a univariate regression. Like influence statistics, leverage plots can be used as a method for identifying influential observations or outliers, as well as a method of graphically diagnosing any potential failures of the underlying assumptions of a regression model.
• The ARMA frequency spectrum view of an ARMA equation shows the spectrum of the estimated ARMA terms in the frequency domain, rather than the typical time domain. Whereas viewing the ARMA terms in the time domain lets you view the autocorrelation functions of the data, viewing them in the frequency domain lets you observe more complicated cyclical characteristics.

TSLS and GMM Diagnostics

• The Instrument Summary view of an equation is available for non-panel equations estimated by GMM, TSLS or LIML. The summary will display the number of instruments specified, the instrument specification, and a list of the instruments that were used in estimation.
• The Instrument Orthogonality test, also known as the C-test or Eichenbaum, Hansen and Singleton (EHS) Test, evaluates the othogonality condition of a sub-set of the instruments. This test is available for non-panel equations estimated by TSLS or GMM.
• The Regressor Endogeneity Test, also known as the Durbin-Wu-Hausman Test, tests for the endogeneity of some, or all, of the equation regressors. This test is available for non-panel equations estimated by TSLS or GMM.
  
  A regressor is endogenous if it is explained by the instruments in the model, whereas exogenous variables are those which are not explained by instruments. In EViews’ TSLS and GMM estimation, exogenous variables may be specified by including a variable as both a regressor and an instrument, whereas endogenous variable are those which are specified in the regressor list only.
• The Weak Instrument Diagnostics view provides diagnostic information on the instruments used during estimation. This information includes the Cragg-Donald statistic, the associated Stock and Yugo critical values, and Moment Selection Criteria (MSC). The Cragg-Donald statistic and its critical values are available for equations estimated by TSLS, GMM or LIML, but the MSC are available for equations estimated by TSLS or GMM only.