Abstract
A novel Bayesian nonparametric test for assessing multivariate normal models is presented. Although there are extensive frequentist and graphical methods for testing multivariate normality, it is challenging to find Bayesian counterparts. The approach considered in this paper is based on the Dirichlet process and the squared radii of observations. Specifically, the squared radii are employed to transform the \(m\)-variate problem into a univariate problem by relying on the fact that if a random sample is coming from a multivariate normal distribution then the square radii follow a particular beta distribution. While the Dirichlet process is used as a prior on the distribution of the square radii, the concentration of the distribution of the Anderson–Darling distance between the posterior process and the beta distribution is compared to that between the prior process and beta distribution via a relative belief ratio. Key results of the approach are derived. The procedure is illustrated through several examples, in which it shows excellent performance.
Similar content being viewed by others
REFERENCES
L. Al-Labadi, Z. Baskurt, and M. Evans, ‘‘Goodness of fit for the logistic regression model using relative belief,’’ Journal of Statistical Distributions and Applications 4, 1 (2017).
L. Al-Labadi, Z. Baskurt, and M. Evans, ‘‘Statistical reasoning: choosing and checking the ingredients, inferences based on a measure of statistical evidence with some applications,’’ Entropy 20, 289 (2018).
L. Al-Labadi and M. Evans, ‘‘Optimal robustness results for relative belief inferences and the relationship to prior-data conflict,’’ Bayesian Analysis 12, 705–728 (2017).
L. Al-Labadi and M. Evans, ‘‘Prior-based model checking,’’ Canadian Journal of Statistics 46, 380–398 (2018).
L. Al-Labadi and M. Zarepour, ‘‘A Bayesian nonparametric goodness of fit test for right censored data based on approximate samples from the beta-Stacy process,’’ The Canadian Journal of Statistics 41, 466–487 (2013).
L. Al-Labadi, and M. Zarepour, ‘‘Goodness of fit tests based on the distance between the Dirichlet process and its base measure,’’ Journal of Nonparametric Statistics 26, 341–357 (2014).
L. Al-Labadi and M. Zarepour, ‘‘Simulations from the Two-Parameter Poisson-Dirichlet Process and the Normalized Inverse-Gaussian Process,’’ Sankhyā A 76, 158–176 (2014).
L. Al-Labadi and M. Zarepour, ‘‘Two-sample Kolmogorov–Smirnov test using a Bayesian nonparametric approach,’’ Mathematical Methods of Statistics 26, 212–225 (2017).
J. A. V. Alva and E. G. Estrada, ‘‘A generalization of Shapiro–Wilk’s test for multivariate normality,’’ Communications in Statistics-Theory and Methods 38, 1870–1883 (2009).
T. W. Anderson and D. A. Darling, ‘‘A test of goodness of fit,’’ Journal of the American Statistical Association 49, 765–769 (1954).
A. C. Atkinson, M. Riani, and A. Cerioli, Exploring Multivariate Data with the Forward Search (Springer, New York, 2004).
A. Batsidis, N. Martin, L. Pardo, and K. Zografos, ‘‘A necessary power divergence type family tests of multivariate normality,’’ Communications in Statistics-Simulation and Computation 42, 2253–2271 (2013).
P. Billingsley, Probability and Measure (Wiley, New York, 1995).
L. Bondesson, ‘‘On simulation from infinitely divisible distributions,’’ Advances in Applied Probability 14, 885–869 (1982).
M. Capiński and E. Kopp, Measure, Integral and Probability, 2nd ed. (Springer, Berlin, 2004).
I. R. Cardoso de Oliveira and D. F. Ferreira, ‘‘Multivariate extension of chi-squared univariate normality test,’’ Journal of Statistical Computation and Simulation 80, 513–526 (2010).
K. Choi and W. G. Bulgren, ‘‘An estimation procedure for mixtures of distributions,’’ Journal of the Royal Statistical Society, B 30, 444–460 (1988).
A. Dasgupta, Asymptotic Theory of Statistics and Probability (Springer, New York, 2008).
J. Doornik and H. Hansen, ‘‘An omnibus test for univariate and multivariate normality,’’ Oxford Bulletin of Economics and Statistics 70, 927–939 (2008).
R. Dubes and A. K. Jain, ‘‘Clustering methodologies in exploratory data analysis,’’ Advances in Computers 19, 113–228 (1980).
M. Evans, ‘‘Bayesian inference procedures derived via the concept of relative surprise,’’ Communications in Statistics-Theory and Methods 26, 1125–1143 (1997).
M. Evans, Measuring Statistical Evidence Using Relative Belief, Vol. 144: Monographs on Statistics and Applied Probability (CRC Press, Boca Raton, 2015).
M. Evans and H. Moshonov, ‘‘Checking for prior-data conflict,’’ Bayesian Analysis 1, 893–914 (2006).
L. Fattorini, ‘‘Remarks on the use of Shapiro–Wilk statistic for testing multivariate normality,’’ Statistica 46, 209–217 (1986).
T. S. Ferguson, ‘‘A Bayesian analysis of some nonparametric problems,’’ The Annals of Statistics 1, 209–230 (1973).
G. Fernandez, Data Mining Using SAS Applications, 2nd ed. (CRC Press, Boca Raton, 2010).
R. Gnanadesikan and J. Kettenring, ‘‘Robust estimates, residuals, and outlier detection with multiresponse data,’’ Biometrics 28, 81–124 (1972).
Z. Hanusz and J. Tarasińska, ‘‘New test for multivariate normality based on Small’s and Srivastava’s graphical methods,’’ Journal of Statistical Computation and Simulation 82, 1743–1752 (2012).
A. M. Hasofer and G. Z. Stein, ‘‘Testing for multivariate normality after coordinate transformation,’’ Communications in Statistics-Theory and Methods 19, 1403–1418 (1990).
M. J. R. Healy, ‘‘Multivariate normal plotting,’’ Applied Statistics 17, 157–161 (1968).
N. Henze and J. Visagie, ‘‘Testing for normality in any dimension based on a partial differential equation involving the moment generating function,’’ Annals of the Institute of Statistical Mathematics (2019). https://doi.org/10.1007/s10463-019-00720-8
N. Henze and B. Zirkler, ‘‘A class of invariant consistent tests for multivariate normality,’’ Communications in Statistics-Theory and Methods 19, 3595–3617 (1990).
H. Holgersson, ‘‘A graphical method for assessing multivariate normality,’’ Computational Statistics 21, 141–149 (2006).
L. F. James, ‘‘Large sample asymptotics for the two-parameter Poisson-Dirichlet process,’’ in: Pushing the Limits of Contemporary Statistics: Contributions in Honor of Jayanta K. Ghosh, eds. B. Clarke and S. Ghosal (Ohio: Institute of Mathematical Statistics, 2008), pp. 187–199.
K. Jönsson, ‘‘A robust test for multivariate normality,’’ Economics Letters 113, 199–201 (2011).
I. Kim and S. Park, ‘‘Likelihood ratio tests for multivariate normality,’’ Communications in Statistics-Theory and Methods 47, 1923,1934 (2018).
N. Kim, ‘‘A robustified Jarque-Bera test for multivariate normality,’’ Economic Letter 140, 48–52 (2016).
M. S. Madukaife and F. C. Okafor, ‘‘A powerful affine invariant test for multivariate normality based on interpoint distances of principal components,’’ Communications in Statistics-Simulation and Computation 47, 1264–1275 (2018).
J. F. Malkovich and A. A. Afifi, ‘‘On tests for multivariate normality,’’ Journal of the American Statistical Association 68, 176–179 (1973).
K. V. Mardia, ‘‘Measures of multivariate skewness and kurtosis with applications,’’ Biometrika 57, 519–530 (1970).
D. J. Nott, M. Seah, L. Al-Labadi, M. Evans, H. K. Ng, and B. Englert, ‘‘Using prior expansions for prior-data conflict checking,’’ Bayesian Analysis (2020). https://projecteuclid.org/euclid.ba/1585360930
R.W. Oldford, ‘‘Self-calibrating quantile-quantile plots,’’ The American Statistician 70, 74–90 (2016).
S. Rincón-Gallardo, C. P. Quesenberry, and F. J. O’Reilly, ‘‘Conditional probability integral transformations and goodness-of-fit tests for multivariate normal distributions,’’ The Annals of Statistics 7, 1052–1057 (1979).
J. P. Royston, ‘‘Some techniques for assessing multivariate normality based on the Shapiro–Wilk W,’’ Applied Statistics 32, 121–133 (1983).
J. Sethuraman, ‘‘A constructive definition of Dirichlet priors,’’ Statistica Sinica 4, 639–650 (1994).
N. Small, ‘‘Plotting squared radii,’’ Biometrika 65, 657–658 (1978).
G. J. Székely and M. L. Rizzo, ‘‘Energy statistics: A class of statistics based on distances,’’ Journal of Statistical Planning and Inference 143, 1249–1272 (2013).
S. T. Tokdar and R. Martin, ‘‘Bayesian test of normality versus a Dirichlet process mixture alternative,’’ Sankhya B 83, 66–96 (2021).
R. L. Wolpert, and K. Ickstadt, Simulation of Lévy random fields. In Practical Nonparametric and Semiparametric Bayesian Statistics, Vol. 133: Lect. Notes Stat. (Springer, New York, 1998), pp. 227–242.
M. Zarepour and L. Al-Labadi, ‘‘On a rapid simulation of the Dirichlet process,’’ Statistics and Probability Letters 82, 916–924 (2012).
M. Zhou and Y. Shao, A powerful test for multivariate normality, Applied Statistics 41, 351–363 (2014).
ACKNOWLEDGMENTS
The authors thank the Editor, Associate Editor, and referees for many helpful comments.
Author information
Authors and Affiliations
Corresponding authors
Appendices
PROOF OF LEMMA 1
Consider \(P_{N}(x)\) as
Let \(g(x)=\frac{dG(x)}{dx}\), then
Substituting \(y=G(x)\), \(G(Y_{(i)})=U_{(i)}\), \(G(-\infty)=0\), and \(G(\infty)=1\), gives
Note that,
Also, \(I_{2}=-U_{(1)}-\log\left(1-U_{(1)}\right)\) and \(I_{3}=-1-\log U_{(N)}+U_{(N)}.\) Therefore, adding \(I_{1}\), \(I_{2}\), and \(I_{3}\), gives
The proof is completed by substituting \(P_{N}(Y_{(i)})=\sum_{j=1}^{i}J^{\prime}_{j}\), \(P^{2}_{N}(Y_{(i)})=\sum_{j=1}^{i}J^{\prime^{2}}_{j}+2\sum_{j=1}^{i-1}\sum_{k=j+1}^{i}J^{\prime}_{k}J^{\prime}_{j}\) and \(U_{(i)}=G(Y_{(i)})\) in terms on the right-hand side of (8).
PROOF OF LEMMA 3
To prove (i), note that, from the property of the Dirichlet process, for any \(t\in\mathbb{R}\), \(E_{P}(P(t)-\) \(H(t))^{2}=\frac{H(t)(1-H(t))}{a+1}\). Then
To prove (ii), it is enough to compute \(E_{P}\left(d_{\textrm{AD}}(P,H)\right)^{2}\). According to the Corollary 2, we consider \(H\) to be the cdf of the Uniform distribution on [0]. Then
Note that, from the property of the Dirichlet process, for any \(s<t\) and \(i,j\in\mathbb{N}\), \(E_{P}(P^{i}(s)P^{j}((s,t]))=\frac{\Gamma(a)}{\Gamma(a+i+j)}\frac{\Gamma(as+i)}{\Gamma(as)}\frac{\Gamma\left(a(t-s)+j\right)}{\Gamma(a(t-s))}\) and \(E_{P}\left(P^{i}(s)\right)=\prod_{k=0}^{i-1}\frac{as+k}{a+k}\). Then
After simplification, we get
for Re\((t)<1\) or \(t\not\in\mathbb{R}\), where Re\((t)\) denotes the real part of \(t\) and \(i\) is the imaginary unit. Then, the variance of \(d_{\textrm{AD}}(P,H)\) is given by
Hence, the proof is completed.
PROOF OF LEMMA 5
Assume that \(r^{\prime}=(R^{\prime}(\mathbf{y}_{1}),\ldots,R^{\prime}(\mathbf{y}_{n}))\) is the observed sample from \(P\) where \(P\sim DP(a,F_{\textrm{beta}})\). Note that
where the third inequality holds since \(0\leq|P_{r^{\prime}}(t)-F_{\textrm{beta}}(t)|\leq 1\) and the fourth inequality holds by triangle inequality. To prove (i), as \(a\rightarrow\infty\), from James [34], \({\sup_{t\in\mathbb{R}}}|P_{r^{\prime}}(t)-H_{r^{\prime}}(t)|\xrightarrow{\mathrm{a.s.}}0\) and by the continuous mapping theorem \({\sup_{t\in\mathbb{R}}}|H_{r^{\prime}}(t)-F_{\textrm{beta}}(t)|\xrightarrow{\mathrm{a.s.}}0\). To prove (ii), since \({\sup_{t\in\mathbb{R}}}|P_{r^{\prime}}(t)-H_{r^{\prime}}(t)|\xrightarrow{\mathrm{a.s.}}0\) as \(n\rightarrow\infty\) and \(\mathcal{H}_{0}\) is true, the continuous mapping theorem and Polya’s theorem [18] imply \({\sup_{t\in\mathbb{R}}}|H_{r^{\prime}}(t)-F_{\textrm{beta}}(t)|\xrightarrow{\mathrm{a.s.}}0\). Note that, the final results of part (i) and (ii) are concluded by practical assumptions in probability and measure theory given in Section 3.1 of Capiński and Kopp [15].
To prove (iii), note that, \(\left(F_{\textrm{beta}}(t)(1-F_{\textrm{beta}}(t))\right)^{-1}\geq 4\), then
From Choi and Bulgren [17], since \(d_{CvM}(P_{r^{\prime}},F_{\textrm{beta}})\geq\dfrac{1}{3}\left(\displaystyle{\sup_{t\in\mathbb{R}}}|P_{r^{\prime}}(t)-F_{\textrm{beta}}(t)|\right)^{3}\),
Using the triangle inequality gives
Similar to the proof of part (ii), as \(n\rightarrow\infty\), \({\sup_{t\in\mathbb{R}}}|P_{r^{\prime}}(t)-H_{(d)}(t)|\xrightarrow{\mathrm{a.s.}}0\) and \({\sup_{t\in\mathbb{R}}}|H_{r^{\prime}}(t)-F_{\textrm{beta}}(t)|\xrightarrow{\mathrm{a.s.}}\displaystyle{\sup_{t\in\mathbb{R}}}|P_{\mathrm{true}}(t)-F_{\textrm{beta}}(t)|\), where \(P_{\mathrm{true}}\) is the true distribution of the sample \(d\). Since \(\mathcal{H}_{0}\) is not true, \(\liminf\displaystyle{\sup_{t\in\mathbb{R}}}|P_{r^{\prime}}(t)-F_{\textrm{beta}}(t)|\displaystyle{\overset{\mathrm{a.s.}}{>}}0\), which implies \(\liminf d_{\textrm{AD}}(P_{r^{\prime}},\) \(F_{\textrm{beta}})\displaystyle{\overset{\mathrm{a.s.}}{>}}0\).
To prove (iv), since for any \(t\in\mathbb{R}\), \(E_{P_{r^{\prime}}}\left(P_{r^{\prime}}(t)\right)=H_{r^{\prime}}(t)\) and \(E_{P_{r^{\prime}}}\left(P_{r^{\prime}}(t)-H_{r^{\prime}}(t)\right)^{2}=\frac{H_{r^{\prime}}(t)\left(1-H_{r^{\prime}}(t)\right)}{a+n+1}\), then, as \(n\rightarrow\infty\)
Applying Fatou’s lemma gives
Since \(\mathcal{H}_{0}\) is not true, \(\inf\left(\frac{\left(P_{\mathrm{true}}(t)-F_{\textrm{beta}}(t)\right)^{2}}{F_{\textrm{beta}}(t)\left(1-F_{\textrm{beta}}(t)\right)}\right)\displaystyle{\overset{\mathrm{a.s.}}{>}}0\). Hence, by Theorem 15.2 of Billingsley [13], \(\liminf E_{P_{r^{\prime}}}\left(d_{\textrm{AD}}(P_{r^{\prime}},F_{\textrm{beta}})\right)\displaystyle{\overset{\mathrm{a.s.}}{>}}0\).
About this article
Cite this article
Al-Labadi, L., Fazeli Asl, F. & Saberi, Z. A Necessary Bayesian Nonparametric Test for Assessing Multivariate Normality. Math. Meth. Stat. 30, 64–81 (2021). https://doi.org/10.3103/S1066530721030029
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S1066530721030029