×

Consistent noisy independent component analysis. (English) Zbl 1429.62215

Summary: We study linear factor models under the assumptions that factors are mutually independent and independent of errors, and errors can be correlated to some extent. Under the factor non-Gaussianity, second-to-fourth-order moments are shown to yield full identification of the matrix of factor loadings. We develop a simple algorithm to estimate the matrix of factor loadings from these moments. We run Monte Carlo simulations and apply our methodology to data on cognitive test scores, and financial data on stock returns.

MSC:

62H25 Factor analysis and principal components; correspondence analysis
62P20 Applications of statistics to economics

Software:

FastICA
PDF BibTeX XML Cite
Full Text: DOI HAL

References:

[1] Albera, L.; Ferreol, A.; Comon, P.; Chevalier, P., Blind identification of overcomplete mixtures of sources (BIOME), Linear algebra and applications, 391, 1-30, (2004) · Zbl 1060.94503
[2] Anderson, T.W., Asymptotic theory for principal component analysis, Annals of mathematical statistics, 34, 122-148, (1963) · Zbl 0202.49504
[3] Anderson, T.W.; Rubin, H., Statistical inference in factor analysis, () · Zbl 0070.14703
[4] Anselin, L., Spatial externalities, spatial multipliers, and spatial econometrics, International regional science review, 26, 153-166, (2003)
[5] Attias, H., Independent factor analysis, Neural computation, 11, 803-851, (1999)
[6] Bai, J., Inferential theory for factor models of large dimensions, Econometrica, 71, 135-171, (2003) · Zbl 1136.62354
[7] Bai, J.; Ng, S., Determining the number of factors in approximate factor models, Econometrica, 70, 191-221, (2002) · Zbl 1103.91399
[8] Beckmann, C.F.; Smith, S.M., Probabilistic independent component analysis for functional magnetic resonance imaging, IEEE transactions on medical imaging, 23, 137-152, (2004)
[9] Blanchard, O.J.; Quah, D., The dynamic effects of aggregate demand and supply disturbances, American economic review, 79, 655-673, (1989)
[10] Bonhomme, S., Robin, J.M., 2008a. Additional appendix to consistent noisy independent component analysis. Available at: http://www.cemfi.es/ bonhomme/ · Zbl 1429.62215
[11] Bonhomme, S., Robin, J.M., 2008b. Generalized nonparametric deconvolution with an application to earnings dynamics. Unpublished working paper · Zbl 1187.62071
[12] Cardoso, J.F., 1991. Super-symmetric decomposition of the fourth-order cumulant tensor. Blind Identification of More Sources Than Sensors. In: Proceedings IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP-91, Toronto, pp. 3109-3112
[13] Cardoso, J.F., High-order contrasts for independent component analysis, Neural computation, 11, 157-192, (1999)
[14] Cardoso, J.F.; Souloumiac, A., Blind beamforming for non-Gaussian signals, IEE-Proceedings-F, 140, 362-370, (1993)
[15] Cardoso, J.F.; Souloumiac, A., Jacobi angles for simultaneous diagonalization, SIAM journal on matrix analysis and applications, 17, 161-164, (1996) · Zbl 0844.65028
[16] Cardoso, J.F., Pham, D.T., 2004. Optimization issues in noisy gaussian ICA. In: Proceedings ICA 2004, Granada, Spain
[17] Carneiro, P.; Hansen, K.; Heckman, J.J., Estimating distributions of treatment effects with an application to the returns to schooling and measurement of the effects of uncertainty on college choice, International economic review, 44, 361-422, (2003)
[18] Chen, A.; Bickel, P.J., Consistent independent component analysis and prewhitening, IEEE transactions on signal processing, 53, 3625-3632, (2005) · Zbl 1373.62292
[19] Chudik, A., Pesaran, M.H., 2007. Infinite dimensional VARs and factor models. Working Paper, University of Cambridge · Zbl 1441.62652
[20] Comon, P., Independent component analysis, a new concept?, Signal processing, 36, 287-314, (1994) · Zbl 0791.62004
[21] Comon, P., Blind identification and source separation in 2×3 under-determined mixtures, IEEE transactions on signal processing, 11-22, (2004) · Zbl 1369.94349
[22] Cragg, J.G., Using higher moments to estimate the simple errors-in-variables model, RAND journal of economics, 28, S71-S91, (1997)
[23] Dagenais, M.G.; Dagenais, D.L., Higher moment estimators for linear regression models with errors in variables, Journal of econometrics, 76, 193-221, (1997) · Zbl 0873.62066
[24] Davies, M., Identifiability issues in noisy ICA, IEEE signal processing letters, 11, 470-473, (2004)
[25] De Lathauwer, L.; Castaing, J.; Cardoso, J.F., Fourth-order cumulant-based blind identification of underdetermined mixtures, IEEE transactions on signal processing, 55, 2965-2972, (2007) · Zbl 1391.94505
[26] De Lathauwer, L.; De Moor, B.; Vandewalle, J., Computation of the canonical decomposition by means of a simultaneous generalized Schur decomposition, SIAM journal of matrix analysis and applications, 26, 295-327, (2004) · Zbl 1080.65031
[27] Erickson, T.; Whited, T., Two-step GMM estimation of the error-in-variables model using high-order moments, Econometric theory, 18, 776-799, (2002) · Zbl 1109.62361
[28] Eriksson, J., Koivunen, V., 2003. Identifiability and separability of linear ICA models revisited. In: 4th International Symposium on ICA and Blind Signal Separation, pp. 23-27
[29] Fama, E.; French, K., Common risk factors in the returns on stocks and bonds, Journal of financial economics, 33, 3-56, (1993) · Zbl 1131.91335
[30] Flury, B., Asymptotic theory for common principal component analysis, Annals of statistics, 14, 418-430, (1986) · Zbl 0613.62075
[31] Forni, M.; Reichlin, L., Let’s get real: A factor analytical approach to disaggregated business cycle dynamics, Review of economic studies, 65, 453-473, (1998) · Zbl 0911.90087
[32] Geary, R.C., Inherent relations between random variables, Proceedings of the royal irish Academy, 47, 63-76, (1942) · Zbl 0063.01553
[33] Heckman, J.J.; Stixrud, J.; Urzua, S., The effect of cognitive and noncognitive abilities on labor market outcomes and social behavior, Journal of labor economics, 24, 411-482, (2006)
[34] Hyvärinen, A.; Oja, E., A fast fixed point algorithm for independent component analysis, Neural computation, 9, 1483-1492, (1997)
[35] Hyvärinen, A.; Karhunen, J.; Oja, E., Independent component analysis, (2001), John Wiley & Sons New York
[36] Ikeda, S.; Toyama, K., Independent component analysis for noisy data—MEG data analysis, Neural networks, 13, 1063-1074, (2000)
[37] Jennrich, R.I.; Trendafilov, N., Independent component analysis as a rotation method: A very different solution to thurstone’s box problem, British journal of mathematical and statistical psychology, 58, 199-208, (2005)
[38] Lewbel, A., Constructing instruments for regressions with measurement error when no additional data are available, with an application to patents and R&D, Econometrica, 65, 1201-1213, (1997) · Zbl 0898.90044
[39] Mencia, J., Sentana, E., 2008. Multivariate location-scale mixtures of normals and mean-variance-skewness portfolio allocation. CEMFI Working Paper 0805 · Zbl 1431.62482
[40] Moulines, E., Cardoso, J.F., Gassiat, E., 1997. Maximum likelihood for blind separation and deconvolution of noisy signals using mixture models. In: Proceedings ICASSP 1997, vol. 5, Munich, pp. 3617-3620
[41] Pal, M., Consistent moment estimators of regression coefficients in the presence of errors-in-variables, Journal of econometrics, 14, 349-364, (1980) · Zbl 0459.62052
[42] Pesaran, M.H., Estimation and inference in large heterogeneous panels with a multifactor error structure, Econometrica, 74, 967-1012, (2006) · Zbl 1152.91718
[43] Reiersol, O., Identifiability of a linear relation between variables which are subject to error, Econometrica, 9, 1-24, (1950)
[44] Robin, J.M.; Smith, R.J., Tests of rank, Econometric theory, 16, 151-175, (2000) · Zbl 0957.62047
[45] Ross, S.A., The arbitrage pricing theory of capital asset pricing, Journal of economic theory, 16, 341-360, (1976)
[46] Schennach, S.M., Hu, Y., Lewbel, A., 2007. Nonparametric identification of the classical errors-in-variables model without side information. Cemmap Working Paper 14/07
[47] Stegeman, A., Mooijaart, A., 2007. Independent factor analysis by least squares. Technical Report, Heymans Institute, University of Groningen. Available online at: http://www.gmw.rug.nl/ stegeman
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.