×

Recognizing and visualizing departures from independence in bivariate data using local Gaussian correlation. (English) Zbl 1322.62140

Summary: It is well known that the traditional Pearson correlation in many cases fails to capture non-linear dependence structures in bivariate data. Other scalar measures capable of capturing non-linear dependence exist. A common disadvantage of such measures, however, is that they cannot distinguish between negative and positive dependence, and typically the alternative hypothesis of the accompanying test of independence is simply “dependence”. This paper discusses how a newly developed local dependence measure, the local Gaussian correlation, can be used to construct local and global tests of independence. A global measure of dependence is constructed by aggregating local Gaussian correlation on subsets of \(\mathbb R^2\), and an accompanying test of independence is proposed. Choice of bandwidth is based on likelihood cross-validation. Properties of this measure and asymptotics of the corresponding estimate are discussed. A bootstrap version of the test is implemented and tried out on both real and simulated data. The performance of the proposed test is compared to the Brownian distance covariance test. Finally, when the hypothesis of independence is rejected, local independence tests are used to investigate the cause of the rejection.

MSC:

62G09 Nonparametric statistical resampling methods
62G10 Nonparametric hypothesis testing
62H20 Measures of association (correlation, canonical correlation, etc.)

Software:

localgauss; sm
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Ansolabehere, S., Snyder, J.M.J., Stewart, C.I.: Candidate positioning in U.S. house elections. Am. J. Polit. Sci. 45, 136-159 (2001) · doi:10.2307/2669364
[2] Berentsen, G.D., Kleppe, T., Tjøstheim, D.: Introducing localgauss, an R package for estimating and visualizing local Gaussian correlation (2013). Working paper. http://people.uib.no/gbe062/local-gaussian-correlation/
[3] Berentsen, G.D., Støve, B., Tjøstheim, D., Nordbø, T.: Recognizing and visualizing copulas: an approach using local Gaussian approximation. Working paper (2013) http://people.uib.no/gbe062/local-gaussian-correlation/ · Zbl 1304.62085
[4] Bickel, J.P., Freedman, D.A.: Some asymptotic theory for the bootstrap. Ann. Math. Stat. 9, 1196-1217 (1981) · Zbl 0449.62034 · doi:10.1214/aos/1176345637
[5] Birgé, B., Massart, P.: Estimation of integral functionals of a density. Ann. Math. Stat. 23, 11-29 (1995) · Zbl 0848.62022 · doi:10.1214/aos/1176324452
[6] Bjerve, S., Doksum, K.: Correlation curves: measures of association as functions of covariate values. Ann. Math. Stat. 21, 890-902 (1993) · Zbl 0817.62025 · doi:10.1214/aos/1176349156
[7] Blum, J.R., Kiefer, J., Rosenblatt, M.: Distribution free tests of independence based on the sample distribution function. Ann. Math. Stat. 32, 485-498 (1961) · Zbl 0139.36301 · doi:10.1214/aoms/1177705055
[8] Bollerslev, T.: Generalized autoregressive conditional heterosked asticity. J. Econom. 31, 307-327 (1986) · Zbl 0616.62119 · doi:10.1016/0304-4076(86)90063-1
[9] Bowman, A., Azzalini, A.: Applied Smoothing Techniques for Data Analysis: The Kernel Approach with S-Plus Illustrations. Oxford Science Publications/Clarendon Press, Oxford (1997) · Zbl 0889.62027
[10] Brock, W., Dechert, W., LeBaron, B., Scheinkman, J.: A test for independence based on the correlation dimension. Econom. Rev. 15, 197-235 (1996) · Zbl 0893.62034 · doi:10.1080/07474939608800353
[11] Chan, N.H., Tran, L.T.: Nonparametric tests for serial dependence. J. Time Ser. Anal. 13, 19-28 (1992) · Zbl 0745.62044 · doi:10.1111/j.1467-9892.1992.tb00092.x
[12] Deheuvels, P.: An asymptotic decomposition for multivariate distribution-free tests of independence. J. Multivar. Anal. 11, 102-113 (1981) · Zbl 0486.62043 · doi:10.1016/0047-259X(81)90136-6
[13] Delicado, P., Smrekar, M.: Measuring non-linear dependence for two random variables distributed along a curve. Stat. Comput. 19, 255-269 (2009). doi:10.1007/s11222-008-9090-y · doi:10.1007/s11222-008-9090-y
[14] Doksum, K., Blyth, S., Bradlow, E., Meng, X., Zhao, H.: Correlation curves as local measures of variance explained by regression. J. Am. Stat. Assoc. 89, 571-582 (1994) · Zbl 0803.62035 · doi:10.1080/01621459.1994.10476782
[15] Forbes, K.J., Rigobon, R.: No contagion, only interdependence: measuring stock market comovements. J. Finance 57, 2223-2261 (2002) · doi:10.1111/0022-1082.00494
[16] Genest, C., Rémillard, B.: Validity of the parametric bootstrap for goodness-of-fit testing in semiparametric models. Ann. Henri Poincaré 44, 1096-1127 (2008) · Zbl 1206.62044 · doi:10.1214/07-AIHP148
[17] Giné, E., Mason, D.: Uniform in bandwidth estimation of integral functionals of the density function. Scand. J. Stat. 35, 739-761 (2008) · Zbl 1194.62029 · doi:10.1111/j.1467-9469.2008.00600.x
[18] Hall, P., Marron, J.P.: Estimation of integrated squared density derivatives. Stat. Probab. Lett. 6, 109-115 (1987) · Zbl 0628.62029 · doi:10.1016/0167-7152(87)90083-6
[19] Hansen, B.: Uniform convergence rates for kernel estimation with dependent data. Econom. Theory 24, 726-748 (2008) · Zbl 1284.62252
[20] Hjort, N.L., Jones, M.C.: Locally parametric nonparametric density estimation. Ann. Stat. 24, 1619-1647 (1996) · Zbl 0867.62030 · doi:10.1214/aos/1032298288
[21] Hoeffding, W.: A non-parametric test of independence. Ann. Math. Stat. 19, 546-557 (1948) · Zbl 0032.42001 · doi:10.1214/aoms/1177730150
[22] Holland, P.W., Wang, Y.J.: Dependence function for continuous bivariate densities. Commun. Stat., Theory Methods 16, 863-876 (1987) · Zbl 0609.62092 · doi:10.1080/03610928708829408
[23] Joe, H.: Estimation of entropy and other functionals of a multivariate density. Ann. Inst. Stat. Math. 41, 683-697 (1989). doi:10.1007/BF00057735 · Zbl 0698.62042 · doi:10.1007/BF00057735
[24] Jones, M.C.: The local dependence function. Biometrika 83, 899-904 (1996) · Zbl 0883.62057 · doi:10.1093/biomet/83.4.899
[25] Jones, M.C.: Constant local dependence. J. Multivar. Anal. 64, 148-155 (1998) · Zbl 0895.62054 · doi:10.1006/jmva.1997.1714
[26] Jones, M.C., Koch, I.: Dependence maps: local dependence in practice. Stat. Comput. 13, 241-255 (2003) · doi:10.1023/A:1024270700807
[27] Lehmann, E.L.: Some concepts of dependence. Ann. Math. Stat. 37, 1137-1153 (1966) · Zbl 0146.40601 · doi:10.1214/aoms/1177699260
[28] Robinson, P.M.: Consistent nonparametric entropy-based testing. Rev. Econ. Stud. 58, 437-453 (1991) · Zbl 0719.62055 · doi:10.2307/2298005
[29] Rosenblatt, M.: A quadratic measure of deviation of two-dimensional density estimates and a test of independence. Ann. Math. Stat. 3, 1-14 (1975) · Zbl 0325.62030 · doi:10.1214/aos/1176342996
[30] Rosenblatt, M., Wahlen, B.E.: A nonparametric measure of independence under a hypothesis of independent components. Stat. Probab. Lett. 15, 245-252 (1992) · Zbl 0770.62039 · doi:10.1016/0167-7152(92)90197-D
[31] Silverman, B.: Density Estimation for Statistics and Data Analysis. Monographs on Statistics and Applied Probability, vol. 26. Chapman & Hall, London (1986) · Zbl 0617.62042 · doi:10.1007/978-1-4899-3324-9
[32] Skaug, H.J.: Nonparametric functional estimation with application to hypothesis testing. Ph.D. thesis, Uib (1994)
[33] Skaug, H. J.; Tjøstheim, D.; Subba Rao, T. (ed.), Nonparametric tests of serial independence, 207-229 (1993), London · Zbl 0880.62052 · doi:10.1007/978-1-4899-4515-0_15
[34] Skaug, H.J., Tjøstheim, D.: A nonparametric test of serial independence based on the empirical distribution function. Biometrika 80, 591-602 (1993b) · Zbl 0790.62044 · doi:10.1093/biomet/80.3.591
[35] Skaug, H.J., Tjøstheim, D.: Testing for serial independence using measures of distance between densities. Lect. Notes Stat. 115, 363-377 (1996) · doi:10.1007/978-1-4612-2412-9_27
[36] Stone, M.: Cross-validatory choice and assessment of statistical predictions (with discussion). J. R. Stat. Soc. B 36, 111-147 (1974) · Zbl 0308.62063
[37] Székely, G.J., Rizzo, M.L.: Brownian distance covariance. Ann. Appl. Stat. 3, 1236-1265 (2009) · Zbl 1196.62077 · doi:10.1214/09-AOAS312
[38] Székely, G.J., Rizzo, M.L., Bakirov, N.K.: Measuring and testing dependence by correlation of distances. Ann. Math. Stat. 35, 2769-2794 (2007) · Zbl 1129.62059 · doi:10.1214/009053607000000505
[39] Teräsvirta, T., Tjøstheim, D., Granger, C.W.J.: Modelling Nonlinear Economic Time Series. Oxford University Press, Oxford (2010) · Zbl 1305.62010 · doi:10.1093/acprof:oso/9780199587148.001.0001
[40] Tjøstheim, D., Hufthammer, K.: Local Gaussian correlation: a new measure of dependence. J. Econom. 172, 33-48 (2013) · Zbl 1443.62288 · doi:10.1016/j.jeconom.2012.08.001
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.