×

Three-way component analysis with smoothness constraints. (English) Zbl 1180.62089

Tucker3 Analysis and CANDECOMP/PARAFAC (CP) are closely related methods for three-way component analysis. Imposing constraints on the Tucker3 or CP solutions can be useful to improve estimation of the model parameters. In the present paper, a method is proposed for applying smoothness constraints on Tucker3 or CP solutions, which is particularly useful in analysing functional three-way data. The usefulness of smoothness constraints on Tucker3 and CP solutions is examined by means of a simulation experiment. Generally, the results of the experiments indicate better estimations of the model parameters. An empirical example illustrates the use of smoothness constraints. The constrained model is more stable and easier to interpret than the unconstrained model.

MSC:

62H25 Factor analysis and principal components; correspondence analysis
65D07 Numerical computation using splines
62H99 Multivariate analysis

Software:

Matlab; fda (R)
PDF BibTeX XML Cite
Full Text: DOI

References:

[1] Alsberg, B. K.; Kvalheim, O. M.: Compression of nth-order data arrays by B-splines. Part 1: theory. J. chemometrics 7, 61-73 (1993)
[2] Andersson, C. A.; Bro, R.: Improving the speed of multi-way algorithms: part I. Tucker3. Chemometrics and intelligent laboratory systems 42, 93-103 (1998)
[3] Bro, R., 1998. Multi-way analysis in the food industry. Models, algorithms and applications. Unpublished Doctoral Thesis, University of Amsterdam, Amsterdam.
[4] Bro, R., Smilde, A.K., 2001. Centering and scaling in component analysis. Submitted for publication.
[5] Browne, M. W.; Cudeck, R.: Alternative ways of assessing model fit. Sociological meth. Res. 21, No. 2, 230-258 (1992)
[6] Bus, A.G., Kroonenberg, P.M., 1982. Reading instruction and learning to read: a longitudinal study. Internat. Report, SOL/82-08. Dept. of Education, University of Groningen, Groningen.
[7] Carroll, J. D.; Chang, J.: Analysis of individual differences in multidimensional scaling via an n-way generalisation of ”Eckart-Young decomposition. Psychometrika 35, 283-319 (1970) · Zbl 0202.19101
[8] Carroll, J. D.; Pruzansky, S.; Kruskal, J. B.: CANDELINC: a general approach to multidimensional analysis of many-way arrays with linear constraints on parameters. Psychometrika 1, 3-24 (1980) · Zbl 0478.62050
[9] De Boor, C.: A practical guide to splines. (1978) · Zbl 0406.41003
[10] Golub, G. H.; Van Loan, C. F.: Matrix computations. (1989) · Zbl 0733.65016
[11] Harshman, R. A.: Foundations of the PARAFAC procedure: models and conditions for an ’exploratory’ multi mode factor analysis. UCLA working papers in phonetics 16, 1-84 (1970)
[12] Harshman, R. A.: Determination and proof of minimum uniqueness conditions for PARAFAC1. UCLA working papers in phonetics 22, 111-117 (1972)
[13] Harshman, R. A.; Lundy, M. E.: Data preprocessing and the extended PARAFAC model. Research methods for multimode data analysis, 216-284 (1984)
[14] Hastie, T. J.; Tibshirani, R. J.: Generalized additive models. (1990) · Zbl 0747.62061
[15] Kiers, H. A. L.: A three-step algorithm for CANDECOMP/PARAFAC analysis of large data sets with multicollinearity. J. chemometrics 12, 155-171 (1998)
[16] Kiers, H. A. L.: Towards a standardized notation and terminology in multiway analysis. J. chemometrics 14, 105-122 (2000)
[17] Kiers, H. A. L.; Harshman, R. A.: Relating two proposed methods for speedup of algorithms for Fitting two- and three-way principal component and related multilinear models. Chemometrics intelligent laboratory systems 36, 31-40 (1997)
[18] Kiers, H. A. L.; Van Mechelen, I.: Three-way component analysis: principles and illustrative application. Psychological meth. 6, 84-110 (2001)
[19] Kroonenberg, P. M.: Three-mode principal component analysis. Theory and applications. (1983) · Zbl 0513.62059
[20] Kroonenberg, P. M.; De Leeuw, J.: Principal component analysis of three-mode data by means of alternating least squares algorithms. Psychometrika 45, 69-97 (1980) · Zbl 0431.62035
[21] Lawson, C. L.; Hanson, R. J.: Solving least squares problems. (1974) · Zbl 0860.65028
[22] Little, R. J. A.; Rubin, D. B.: Statistical analysis with missing data. (1987) · Zbl 0665.62004
[23] MATLAB, 2001. MATLAB. The Language of Technical Computing. The Mathworks, Inc., Natick, MA.
[24] Mitchell, B. C.; Burdick, D. S.: Slowly converging PARAFAC sequences: swamps and two-factor degeneracies. J. chemometrics 8, 155-168 (1994)
[25] Ramsay, J. O.: Monotone regression splines in action. Statist. sci. 3, No. 4, 425-461 (1988)
[26] Ramsay, J. O.; Silverman, B. W.: Functional data analysis. (1997) · Zbl 0882.62002
[27] Rao, C. R.: Some statistical methods for comparison of growth curves. Biometrics 14, 1-17 (1958) · Zbl 0079.35704
[28] Stevens, J.: Applied multivariate statistics for the social sciences. (1992) · Zbl 0706.62003
[29] Tucker, L.R., 1951. A method for synthesis of factor analysis studies. Personnel Research Section Report No. 984. Dept. of the Army, Washington D.C.
[30] Tucker, L. R.: Determination of parameters of a functional relationship by factor analysis. Psychometrika 23, 19-23 (1958) · Zbl 0086.13404
[31] Tucker, L. R.: Some mathematical notes on three-mode factor analysis. Psychometrika 31, 279-311 (1966)
[32] Van Rijckevorsel, J. L. A.: Fuzzy coding and B-splines. Component and correspondence analysis. Dimension reduction by functional approximation (1988)
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. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.