Abstract
Disjoint factor analysis (DFA) is a new latent factor model that we propose here to identify factors that relate to disjoint subsets of variables, thus simplifying the loading matrix structure. Similarly to exploratory factor analysis (EFA), the DFA does not hypothesize prior information on the number of factors and on the relevant relations between variables and factors. In DFA the population variance–covariance structure is hypothesized block diagonal after the proper permutation of variables and estimated by Maximum Likelihood, using an Coordinate Descent type algorithm. Inference on parameters on the number of factors and to confirm the hypothesized simple structure are provided. Properties such as scale equivariance, uniqueness, optimal simplification of loadings are satisfied by DFA. Relevant cross-loadings are also estimated in case they are detected from the best DFA solution. DFA has also the option to constrain a variable to load on a pre-specified factor so that the researcher can assume, a priori, some relations between variables and loadings. A simulation study shows performances of DFA and an application to optimally identify the dimensions of well-being is used to illustrate characteristics of the new methodology. A final discussion concludes the paper.
Similar content being viewed by others
Notes
For matrix V columns and rows will be denoted by the \(\mathbf{v}_{.h}\) and \(\mathbf{v}_j\)., respectively.
The OECD defines regions as the first tier of sub-national government (for example states in the United States, provinces in Canada, “régions” in France, “regioni” in Italy).
The Cronbach’s Alpha measures an internal consistency: \(\upalpha \ge 0.9, excellent; 0.7 \le \upalpha < 0.9\), good; \(0.6 \le \upalpha < 0.7\), acceptable; \(0.5 \le \upalpha < 0.6\) poor; \(\upalpha < 0.5\), Unacceptable.
Note that in 50 random starts the same optimal solution was found 41 times.
Note that in 50 random starts the optimal solution was found 40 times.
References
Abdi H (2003) Factor Rotations in Factor Analyses. In: Lewis-Beck M, Bryman A, Futing T (eds) Encyclopedia of Social Sciences Research Methods, pp 792–795. Sage, Thousand Oaks (CA)
Amstrong JS, Soelberg P (1968) On the interpretation of factor analysis. Psychol Bull 70(5):361–364
Akaike H (1973) Information theory and an extension of the maximum likelihood principle. In: Petrov BN, Csáki F (eds) 2nd International Symposium on Information Theory, Tsahkadsor, Armenia, USSR. Akadémiai Kiadó, Budapest, pp 267–281
Anderson TW, Rubin H (1956) Statistical inferences in factor analysis. In: Proceedings of the Third Symposium on Mathematical Statistics and Probability, vol V, pp 111–150
Aoyagi H, Ogunseitan OA (2015) Toxic releases and risk disparity: a spatiotemporal model of industrial ecology and social empowerment. Int J Environ Res Public Health 12:6300–6318
Bartlett MS (1937) The statistical conception of mental factors. Br J Psychol 28:97–104
Bartlett MS (1954) A note on the multiplying factors for various \(\chi \)2 approximations. J R Stat Soc Ser B (Methodol) 16(2):296–298
Boer D, Abubakar A (2014) Music listening in families and peer groups: benefits for young people’s social cohesion and emotional well-being across four cultures. Frontiers Psychol 5(392):1–15
Costello AB, Osborne JW (2005) Best practices in exploratory factor analysis: four recommendations for getting the most from your analysis. Pract Assess Res Eval. http://pareonline.net/getvn.asp?v=10&n=7
d’Aspremont A, El Ghaoui L, Jordan MI, Lanckriet GRG (2007) A direct formulation for sparse PCA using semidefinite programming. SIAM Rev 49(3):434–448
Elliot AJ, McGregor HA (2001) A 2 \(\times \) 2 achievement goal Framework. J Pers Soc Psychol 80(3):501–519
Grbovic M, Dance CR, Vucetic S (2012) Sparse principal component analysis with constraints. In: Proceedings of the twenty-sixth AAAI conference on artificial intelligence. https://www.aaai.org/ocs/index.php/AAAI/AAAI12/paper/viewFile/5109/5503
Horst P (1965) Factor analysis of data matrices. Holt, Rinehart and Winston, New York
Hubert L, Arabie P (1985) Comparing partitions. J Classif 2:193–218
Joereskog KG (1969) A general approach to confirmatory maximum-likelihood factor analysis. Psychometrika 34:183–202
Joereskog KG (1978) Structural analysis of covariance and correlation matrices. Psychometrika 43:443–477
Joereskog KG (1979) A general approach to confirmatory maximum likelihood factor analysis, with addendum. In: Joereskog KG, Sorbom D, Magidson J (eds) Advances in factor analysis and structural equation models, pp 40–43
Joereskog KG, Sorbom D (1979) Advances in factor analysis and equation models. Abt Books, Cambridge
Joereskog KG, Sorbom D (1981) LISREL V: Analysis of linear structural relationships by the method of maximum likelihood [Computer program]. Scientific Software, Mooresville
Jolliffe IT, Trendafilov NT, Uddin M (2003) A modified principal component technique based on the LASSO. J Comput Graph Stat 12(3):531–547
Kaiser HF (1958) The varimax criterion for analytic rotation in factor analysis. Psychometrika 23:187–200
Kim EJ (2013) Accessing factor structure and construct validity of the successful aging inventory. J Korean Acad Nurs 43(4):568–578
Kline P (2000) The handbook of psychological testing (2nd edn.). Routledge, London, p 13
Leiva-Bianchi MC, Araneda AC (2013) Validation of the Davidson Trauma Scale in its original and a new shorter version in people exposed to the F-27 earthquake in Chile. Eur J Psychotraumatology 4:21239
Masaya I, Satomi N, Daisuke F, Mitsunori M, Kim Y, Shear MK, Angela G, Wall MM (2012) Brief measure for screening complicated grief: reliability and discriminant validity. PLoS One 7(2):e31209
Morikawa M, Okada T, Ando M, Aleksic B, Kunimoto S, Nakamura Y, Kubota C,, Uno Y, Tamaji A, Hayakawa N, Furumura K, Shiino T, Morita T, Ishikawa N, Ohoka H, Usui H, Banno N, Murase S, Goto S, Kanai A, Masuda T, Ozaki1 N (2015) Relationship between social support during pregnancy and postpartum depressive state: a prospective cohort study. Nature 5:10520
Mulaik SA (1972) The foundations of factor analysis. McGraw Hill, New York
Park Y-M, Kwon HS, Lim SY, Lee JH, Kim SR, Yoon KH, Cha BY, Son HY, Park YG, Kim DS, Meng K, Le WC (2006) Clustering characteristics of risk variables of metabolic syndrome in Korean rural populations. J Korean Diabetes Assoc 30:177–189
Raftery AE (1993) Bayesian model selection in structural equation models. In: Bollen KA, Long JS (eds) Testing structural equation models. sage, Newbury Park, pp 163–180
Rosman A, Maggie G, Lee S, Butler S, Schwartzy A (2013) DOSPERT\(+\)M: A survey of medical risk attitudes in the United States. Judgm Decis Mak 8(4):470–481
Sherin RJ (1966) A matrix formulation of Kaiser’s varimax criterion. Psychometrika 31(4):535–538
Shen H, Huang JZ (2008) Sparse principal component analysis via regularized low rank matrix approximation. J Multivar Anal 99:1015–1034
Spearman C (1904) General intelligence, objectively determined and measured. Am J Psychol 15:201–293
Steiger JH, Lind J (1980) Statistically based tests for the number of common factors. Paper presented at the annual meeting of the Psychometric Society, Iowa City
Steiger JH (1989) EzPATH: A supplementary module for SYSTAT and SYSGRAPH [Computer program]. SYSTAT Inc, Evanston
Stiglitz JE, Sen A, Fitoussi JP (2009) Report by the Commission on the Measurement of Economic Performance and Social Progress. Insee. http://www.insee.fr/fr/publications-et-services/dossiers_web/stiglitz/doc-commission/RAPPORT_anglais.pdf
Schwarz G (1978) Estimating the dimension of a model. Ann Stat 6(2):461–464
ten Berge JMF (1984) A joint treatment of varimax rotation and the problem of diagonalizing symmetric matrices simultaneously in the least-squares sense. Psychometrika 49(3):347–358
ten Berge JMF, Knol DL, Kiers HAL (1988) A treatment of the orthomax rotation family in terms of diagonalization, and a re-examination of a singular value approach to varimax rotation. Comput Stat Q 3:207–217
Thomson GH (1935) The definition and measurement of g (general intelligence). J Educ Psychol 26:241–262
Ullman JB (2006) Structural equation modeling: reviewing the basics and moving forward. J Personal Assess 87(1):35–50
Vichi M, Saporta G (2009) Clustering and disjoint principal component analysis. Comput Stat Data Anal 53(8):3194–3208
Yang B, Fu X, Sidiropoulos ND (2015) Joint factor analysis and latent clustering. In: Proceedings of IEEE 6th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP2015), pp 173–176
Zangwill WI (1969) Nonlinear programming: a unified approach. Prentice-Hall, Englewood Cliffs
Zou H, Hastie T, Tibshirani R (2006) Sparse principal component analysis. J Comput Graph Stat 15:262–286
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Vichi, M. Disjoint factor analysis with cross-loadings. Adv Data Anal Classif 11, 563–591 (2017). https://doi.org/10.1007/s11634-016-0263-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11634-016-0263-9