Abstract
In applications of canonical correlation analysis (CCA), rotation of the canonical loadings is recommended to facilitate the interpretation of canonical variates. Based on the COSAN modeling approach to CCA proposed by Gu et al. (Multivar Behav Res 54:192–223, 2019), we describe the infinitesimal jackknife (IJ) method in modified COSAN-CCA models to obtain the IJ estimates of standard errors for rotated CCA estimates. Specifically, given two CCA rotation strategies (i.e., concurrent and separate) and two types of rotation method (i.e., orthogonal and oblique), our descriptions of the modified COSAN-CCA models and IJ method cover four rotation strategy-method combinations: concurrent-orthogonal, concurrent-oblique, separate-orthogonal, and separate-oblique. Simulation studies are used to evaluate the IJ estimates of standard errors, and a real data example is analyzed for illustration. We conclude that the IJ method is a trustworthy method for standard error estimation for rotated CCA estimates.
This is a preview of subscription content, log in via an institution to check access.
Notes
On minor disadvantage of the IJ method is that one must have access to the raw data, which may not be available from the literature.
References
Anderson TW (1999) Asymptotic theory for canonical correlation analysis. J Multivar Anal 70:1–29. https://doi.org/10.1006/jmva.1999.1810
Archer CO, Jennrich RI (1973) Standard errors for rotated factor loadings. Psychometrika 38:581–592. https://doi.org/10.1007/BF02291496
Bagozzi RP, Fornell C, Larcker DF (1981) Canonical correlation analysis as a special case of a structural relations model. Multivar Behav Res 16:437–454. https://doi.org/10.1207/s15327906mbr1604_2
Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Roy Stat Soc B 57:289–300. https://doi.org/10.1111/j.2517-6161.1995.tb02031.x
Benjamini Y, Yehutieli D (2005) False discovery rate adjusted multiple confidence intervals for selected parameters. J Am Stat Assoc 100:71–93. https://doi.org/10.1198/016214504000001907
Bentler PM, Huba GJ (1982) Symmetric and asymmetric rotations in canonical correlation analysis: new methods with drug variable examples. In: Hirschberg N, Humphreys LG (eds) Multivariate applications in the social sciences. Erlbaum, Hillsdale, pp 21–46
Browne MW (1984) Asymptotic distribution-free methods for the analysis of covariance structures. Br J Math Stat Psychol 37:62–83. https://doi.org/10.1111/j.2044-8317.1984.tb00789.x
Browne MW (2001) An overview of analytic rotation in exploratory factor analysis. Multivar Behav Res 36:111–150. https://doi.org/10.1207/S15327906MBR3601_05
Clarkson DB (1979) Estimating the standard errors of rotated factor loadings by jackknifing. Psychometrika 44:297–313. https://doi.org/10.1007/BF02294695
Cliff N, Krus DJ (1976) Interpretation of canonical analysis: rotated vs. unrotated solutions. Psychometrika 41:35–42. https://doi.org/10.1007/BF02291696
Crawford CB, Ferguson GA (1970) A general rotation criterion and its use in orthogonal rotation. Psychometrika 35:321–332. https://doi.org/10.1007/BF02310792
Cudeck R (1989) Analysis of correlation matrices using covariance structure models. Psychol Bull 105:317–327. https://doi.org/10.1037/0033-2909.105.2.317
Gu F, Wu H (2018) Simultaneous canonical correlation analysis with invariant canonical loadings. Behaviormetrika 45:111–132. https://doi.org/10.1007/s41237-017-0042-8
Gu F, Yung Y-F, Cheung MW-L (2019) Four covariance structure models for canonical correlation analysis: a COSAN modeling approach. Multivar Behav Res 54:192–223. https://doi.org/10.1080/00273171.2018.1512847
Hotelling H (1935) The most predictable criterion. J Educ Psychol 26:139–142. https://doi.org/10.1037/h0058165
Hotelling H (1936) Relations between two sets of variates. Biometrika 28:321–377. https://doi.org/10.2307/2333955
Jennrich RI (1973) Standard errors for obliquely rotated factor loadings. Psychometrika 38:593–604. https://doi.org/10.1007/BF02291497
Jennrich RI (1974) Simplified formulae for standard errors in maximum-likelihood factor analysis. Br J Math Stat Psychol 27:122–131. https://doi.org/10.1111/j.2044-8317.1974.tb00533.x
Jennrich RI (2008) Nonparametric estimation of standard errors in covariance analysis using the infinitesimal jackknife. Psychometrika 73:579–594. https://doi.org/10.1007/s11336-008-9083-y
Jennrich RI, Clarkson DB (1980) A feasible method for standard errors of estimate in maximum likelihood factor analysis. Psychometrika 45:237–247. https://doi.org/10.1007/BF02294078
Jöreskog KG (1970) A general method for analysis of covariance structures. Biometrika 57:239–251. https://doi.org/10.2307/2334833
Jöreskog KG, Goldberger AS (1975) Estimation of a model with multiple indicators and multiple causes of a single latent variable. J Am Stat Assoc 7:631–639. https://doi.org/10.2307/2285946
Kaiser HF (1958) The varimax criterion for analytic rotation in factor analysis. Psychometrika 23:187–200. https://doi.org/10.1007/BF02289233
Lambert ZV, Wildt AR, Durand RM (1988) Redundancy analysis: An alternative to canonical correlation and multivariate multiple regression in exploring interset associations. Psychol Bull 104:282–289. https://doi.org/10.1037/0033-2909.104.2.282
Marcoulides GA, Hershberger SL (1997) Multivariate statistical methods: a first course. Lawrence Erlbaum Associates Inc., New Jersey
McDonald RP (1978) A simple comprehensive model for the analysis of covariance structures. Br J Math Stat Psychol 31:59–72. https://doi.org/10.1111/j.2044-8317.1978.tb00573.x
McDonald RP (1980) A simple comprehensive model for the analysis of covariance structures: some remarks on applications. Br J Math Stat Psychol 33:161–183. https://doi.org/10.1111/j.2044-8317.1980.tb00606.x
Ogasawara H (2007) Asymptotic expansions of the distributions of estimators in canonical correlation analysis under nonnormality. J Multivar Anal 98:1726–1750. https://doi.org/10.1016/j.jmva.2006.12.001
Pennell R (1972) Routinely computable confidence intervals for factor loadings using the ‘jack-knife.’ Br J Math Stat Psychol 25:107–114. https://doi.org/10.1111/j.2044-8317.1972.tb00481.x
Perreault WD, Spiro RL (1978) An approach for improved interpretation of multivariate analysis. Decis Sci 9:402–413. https://doi.org/10.1111/j.1540-5915.1978.tb00729.x
Satomura H, Adachi K (2013) Oblique rotation in canonical correlation analysis reformulated as maximizing the generalized coefficient of determination. Psychometrika 78:526–537. https://doi.org/10.1007/S11336-012-9310-4
Satorra A, Bentler PM (1994) Corrections to test statistics and standard errors in covariance structure analysis. In: von Eye A, Clogg CC (eds) Latent variables analysis: applications for development research. Sage, Newbury Park, pp 399–419
Thorndike RM (1976) Strategies for rotating canonical components. In: Paper presented to the annual meeting of the American Educational Research Association, San Francisco, CA. (ERIC Document Reproduction Service, ED 123 259)
Thorndike RM (2000) Canonical correlation analysis. In: Tinsley HEA, Brown SD (eds) Handbook of applied multivariate statistics and mathematical modeling. Academic Press, San Diego, pp 237–263
Wilkins JLM (2016) An assessment of the quantitative literacy of undergraduate students. J Exp Educ 84:639–665. https://doi.org/10.1080/00220973.2015.1111854
Wu H, Neale MC (2013) On the likelihood ratio tests in bivariate ACDE models. Psychometrika 78:441–463. https://doi.org/10.1007/s11336-012-9304-2
Yamashita N (2012) Canonical correlation analysis formulated as maximizing sum of squared correlations and rotation of structure matrices. Jpn J Behav 39:1–9. https://doi.org/10.2333/jbhmk.39.1 (in Japanese)
Zhang G, Preacher KJ, Luo S (2010) Bootstrap confidence intervals for ordinary least squares factor loadings and correlations in exploratory factor analysis. Multivar Behav Res 45:104–134. https://doi.org/10.1080/00273170903504836
Zhang G, Preacher KJ, Jennrich RI (2012) The infinitesimal jackknife with exploratory factor analysis. Psychometrika 77:634–648. https://doi.org/10.1007/S11336-012-9281-5
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
About this article
Cite this article
Gu, F., Wu, H., Yung, YF. et al. Standard error estimates for rotated estimates of canonical correlation analysis: an implementation of the infinitesimal jackknife method. Behaviormetrika 48, 143–168 (2021). https://doi.org/10.1007/s41237-020-00123-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s41237-020-00123-7