You are on page 1of 5

http://www.eviews.com/EViews7/ev7eccomp_n.

html

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: inear, og- inear, the !atmull-"om #pline, and the !ardinal #pline.

Whitening
EViews now offers eas$-to-use tools for whitening a series or group of series using A" or VA" regressions, respectivel$. %hitening can be performed with or without a constant and row weights, using a fixed or infocriterion based lag selection. &he coefficients of the whitening regression ma$ be saved.

Long-run Covariances
'ou ma$ now compute estimates of the long-run variance of a series or the long-run covariance matrix of a group of series. 'ou will find this feature in the View menu of a series or a group ob(ect.

EViews provides powerful, eas$-to-use tools for computing, displa$ing, and saving the long-run covariance )variance* matrix of a single series or all of the series in a group ob(ect. 'ou ma$ compute s$mmetric or one-sided long-run covariances using nonparametric +ernel ),ewe$-%est -./7, Andrews -..-*, parametric VA"0A! )1en 0aan and evin -..7*, and prewhitened +ernel )Andrews and 2onahan -..3* methods. 4n addition, EViews supports Andrews )-..-* and ,ewe$-%est )-..5* automatic bandwidth selection methods for +ernel estimators, and information criteria based lag length selection methods for VA"0A! and prewhitening estimation.

6$ default, EViews will estimate the s$mmetric long-run covariance matrix using a non-parametric +ernel estimator with a 6artlett +ernel and a real-valued bandwidth determined solel$ using the number of observations. &he data will be centered )b$ subtracting off means* prior to computing the +ernel covariance estimator, but no other pre-whitening will be performed. &he results will onl$ be displa$ed in the series or group window. 'ou ma$ use the dialog to change these settings.

Variance Ratio Testing


EViews 7 now has built-in variance ratio testing. &he variance ratio test view allows $ou to perform the o and 2ac7inla$ variance ratio test to determine whether differences in a series are uncorrelated, or follow a random wal+ or martingale propert$.

EViews provides a range of testing options. 'ou ma$ perform the o and 2ac7inla$ variance ratio test for homos+edastic and heteros+edastic random wal+s, using the as$mptotic normal distribution ) o and 2ac7inla$, -.//* or wild bootstrap )7im, 3889* to evaluate statistical significance. 4n addition, $ou ma$ compute the ran+, ran+-score, or sign-based forms of the test )%right, 3888*, with bootstrap evaluation of significance. 4n addition, EViews offers %ald and multiple comparison variance ratio tests )"ichardson and #mith, -..-: !how and 1enning, -..;*, so $ou ma$ perform (oint tests of the variance ratio restriction for several intervals.

Cointegration Tests
&o supplement the existing <ohansen cointegration tests, EViews 7 offers support for Engle and =ranger )-./7* and >hillips and ?uliaris )-..8* residual-based tests, 0ansen@s )-..3b* instabilit$ test, and >ar+@s )-..3* added variables test.

&he residual based tests ma$ be computed as a View of a =roup ob(ect, or as a diagnostic view for an equation estimated using one of the cointegrating regression techniques.