An Omnibus Test for Univariate and Multivariate Normality

By Jurgen A. Doornik, Henrik Hansen. 2008

1,057 citations recorded by the source corpus.

Subfields: Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/c3f2f2505eafd326901cc78d93f3694d0f8e694b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Evaluating the Calibration of Multi-Step-Ahead Density Forecasts Using Raw Moments (2015, background): “For more details see, for example, Doornik and Hansen (2008) and the references therein.”
Global Validation of Linear Model Assumptions (2006, result): “The rate of convergence to the χ21 distribution for this statistic is rather slow, as has been noted in earlier articles (see, e.g., Doornik and Hansen 1994).”
Stakeholder Pressure on MNEs and the Transfer of Socially Irresponsible Practices to Subsidiaries (2013, result): “The Doornik and Hansen’s (2008) test for the multivariate normality of these nondummy variables supported the findings of univariate statistics.”
Trend–Cycle Decompositions with Correlated Components (2006, methodology): “Q (12) is the Ljung-Box portmanteau test statistic for residual autocorrelation based on the first 12 autocorrelations, and we also present the Doornik and Hansen (1994) test of normality.”
Fish Price Volatility (2014, background): “We also note that specification tests show no evidence of heteroskedasticity or non-normality Note: Normality refers to the Doornik and Hansen (2008) skewness and kurtosis normality test.”
Multivariate singular spectrum analysis for forecasting revisions to real-time data (2011, methodology): “To assess the normality aspect of our data set, we used the Doornik-Hansen Omnibus, DHO, multivariate normality test 7 , see Doornik and Hansen (1994). This is a multivariate version of Shenton and Bowman‟s (1977) univariate omnibus test for normality, based on transformed skewness and kurtosis coef”
Distributional Tests in Multivariate Dynamic Models with Normal and Student-t Innovations (2009, positive): “Finally, note that both LMk(θ̃T ) and LMs(θ̃T ) are again numerically invariant to the way in which the conditional covariance matrix is factorized, unlike the statistics proposed by Lütkephohl (1993), Doornik and Hansen (1994), and Kilian and Demiroglu (2000), who apply univariate Jarque and Bera (”
Asymmetric causality tests with an application (2012, methodology): “Cumulative positive and negative shocks are used TheoptimallagorderintheVARmodelwasselectedbasedontheminimizationoftheinformationcriterion expressed by Eq. 4 The test suggested by Doornik and Hansen (2008) was used for testing the null hypothesis of multivariate normality A bootstrap multivariate ”
Capturing the Zero: A New Class of Zero-Augmented Distributions and Multiplicative Error Processes (2010, neutral): “The ACM residuals are defined as: ut = It−π̂t√ π̂t (1−π̂t) . consider the omnibus tests proposed by Bowman and Shenton (1975) and Doornik and Hansen (2008), which will be referred to as BS and DH test, respectively.”
Information and Efficiency: Goal Arrival in Soccer Betting (2014, background): “In order to provide assurance that the leptokurtic assumption is appropriate, we provide in Table A.4 the results of the Doornik and Hansen (2008) test for normality.”
Method for the subjective assessment of intermediate quality level of (2014, methodology): “3 It is generally advised to select the most powerful statistical analysis method that is permitted by the data [9] [10].”
Optimal Portfolio Allocation under Higher Moments (2006, neutral): “(2) The omnibus statistic (Omnibus) proposed by Doornik and Hansen (1994) is based on the approximated nite-sample distribution of skewness and kurtosis.”
On tests for multivariate normality and associated simulation studies (2007, background): “Statistics and Computing, 2, 117–119.] performs effectively at detecting departures from MVN for smaller sample sizes.”