## Computation

**Interpolation tools**(Linear, Log-linear, Catmul-Rom Spline, Cardinal Spline) are now offered as a series procedure.**Whitening**is now offered as a series or group procedure.**Long-run variances and covariances**may now be calculated from a series or group of series.**Variance ratio testing**for a random walk or martingale in a series.**Single equation tests for cointegration**are provided for a group of series or for an equation estimated using cointegrating regression.

## Estimation

- Greatly improved single equation
**Instrumental Variables/Two-Stage Least Squares****and GMM estimation**(with new support for LIML and K-class estimation). - Bult-in support for specification and estimation of single equation
**cointegrating regression**. - Support for specification and estimation of
**Generalized Linear Models**. - New methods of
**specifying weights for weighted least squares**.

## Diagnostics

**Expanded choice of coefficient covariance estimators**for single equation regression models.**Enhanced post-estimation diagnostics**for single equation regression models.**New test and diagnostic views**for TSLS and GMM equations.

## EViews 7 New Econometrics and Statistics: Computation

EViews 7 features a number of additions and improvements to its toolbox of basic statistical procedures. Among the highlights are new tools for interpolation, whitening regression, long-run covariance calculation, variance ratio testing, and single-equation cointegration testing.## Interpolation

EViews 7 now offers built-in interpolation series to fill in missing values within a series. EViews offers a number of different algorithms for performing the interpolation: Linear, Log-Linear, the Catmull-Rom Spline, and the Cardinal Spline.

## Whitening

EViews now offers easy-to-use tools for whitening a series or group of series using AR or VAR regressions, respectively. Whitening can be performed with or without a constant and row weights, using a fixed or info-criterion based lag selection. The coefficients of the whitening regression may be saved.

## Long-run Covariances

You may now compute estimates of the long-run variance of a series or the long-run covariance matrix of a group of series. You will find this feature in the View menu of a series or a group object.EViews provides powerful, easy-to-use tools for computing, displaying, and saving the long-run covariance (variance) matrix of a single series or all of the series in a group object. 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.

By default, EViews will estimate the symmetric long-run covariance matrix using a non-parametric kernel estimator with a Bartlett kernel and a real-valued bandwidth determined solely using the number of observations. The data will be centered (by subtracting off means) prior to computing the kernel covariance estimator, but no other pre-whitening will be performed. The results will only be displayed in the series or group window. You may use the dialog to change these settings.

## Variance Ratio Testing

EViews 7 now has built-in variance ratio testing. The variance ratio test view allows you to perform the Lo and MacKinlay variance ratio test to determine whether differences in a series are uncorrelated, or follow a random walk or martingale property.EViews provides a range of testing options. You may perform the Lo and MacKinlay variance ratio test for homoskedastic and heteroskedastic random walks, using the asymptotic normal distribution (Lo and MacKinlay, 1988) or wild bootstrap (Kim, 2006) to evaluate statistical significance. In addition, you may compute the rank, rank-score, or sign-based forms of the test (Wright, 2000), with bootstrap evaluation of significance. In addition, EViews offers Wald and multiple comparison variance ratio tests (Richardson and Smith, 1991; Chow and Denning, 1993), so you may perform joint tests of the variance ratio restriction for several intervals.

## Cointegration Tests

To supplement the existing Johansen cointegration tests, EViews 7 offers support for Engle and Granger (1987) and Phillips and Ouliaris (1990) residual-based tests, Hansen’s (1992b) instability test, and Park’s (1992) added variables test.The residual based tests may be computed as a View of a Group object, or as a diagnostic view for an equation estimated using one of the cointegrating regression techniques.

## EViews 7 New Econometrics and Statistics: Estimation

EViews 7 new estimation features include improved IV and GMM estimation, sophisticated tools for performing cointegrating regression, and estimation of Generalized Linear Models.## Instrumental Variables and GMM Estimation

The algorithms for Instrumental Variables/Two-stage Least Squares estimation of models specified by expression with AR terms have been improved significantly. Limited Information Maximum Likelihood (LIML) and K-class estimation is now available as a single equation estimation method. New options allow you to choose from an expanded set of robust standard error calculations and to not include the constant as an instrument in TSLS.Single equation GMM has been completely overhauled. There is an expanded set of options for the HAC weighting matrix (nonparametric kernel (Newey-West 1987, Andrews 1991), parametric VARHAC (Den Haan and Levin 1997), and prewhitened kernel (Andrews and Monahan 1992) methods, 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 ability to not include a constant as an instrument, the ability to estimate via continuously updating estimation (CUE), and a host of new standard error options, including Windmeijer standard errors. You may now specify prior observation weights.

GMM also offers the ability to save the weighting matrix from estimation and standard error computation, or to use a user-supplied weighting matrix as part of estimation. These features allow the user to estimate a GMM model using the weighting matrix from a previously estimated GMM model.

All three types of IV estimation offer new diagnostics and tests, including a Instrument Orthogonality Test, a Regressor Endogeneity Test, a Weak Instrument Test, and a GMM specific breakpoint test.

## Cointegrating Regression

In addition to the previously supported Johansen system methodology, EViews 7 offers a full set of tools for estimating and testing single equation cointegrating relationships. Three fully efficient estimation methods, Fully Modified OLS (Phillips and Hansen 1992), Canonical Cointegrating Regression (Park 1992), and Dynamic OLS (Saikkonen 1992, Stock and Watson 1993) are described, along with various cointegration testing procedures: Engle and Granger (1987) and Phillips and Ouliaris (1990) residual-based tests, Hansen's (1992b) instability test, and Park's (1992) added variables test.

## Generalized Linear Models

EViews 7 supports estimation of Generalized Linear Models (Nelder and McCullagh, 1983). This class of models generalizes classical linear regression to include a broad range of specifications that have proven to be useful in practice. Among these models are log-linear regression, standard probit and logit, probit and logit specified by proportions, and regression with count or survival data.

A wide range of family, link, dispersion estimation, and estimation options are offered, allowing for computation of various robust standard error and QMLE specifications.

Notably, EViews estimates both prior variance and frequency weighted specifications.

## Weighted Least Squares

The specification of weights in Weighted Least Squares has been generalized so that you may now provide your weights in inverse variance, standard deviation, or variance form. Previously weights were only specified in inverse standard deviation form. Additionally, you may now control whether or not to scale the weight series prior to use. Together, these options should make it easier to match intermediate calculations and results of other sources.## 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.