Abstract.
We present a general method to detect and extract from a finite time sample statistically meaningful correlations between input and output variables of large dimensionality. Our central result is derived from the theory of free random matrices, and gives an explicit expression for the interval where singular values are expected in the absence of any true correlations between the variables under study. Our result can be seen as the natural generalization of the Marčenko-Pastur distribution for the case of rectangular correlation matrices. We illustrate the interest of our method on a set of macroeconomic time series.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
For a review, see: J.H. Stock, M.W. Watson, Implications of dynamical factor models for VAR analysis, working paper, June 2005
C.W.J. Granger, J. Econometrics 100, 17 (2001)
J. Geweke, The Dynamic Factor Analysis of Economic Time Series, Latent Variables in Social Economic Models edited by D.J. Aigner, A.S. Goldberger (North Holland, Amsterdam, 1997)
J.H. Stock, M.W. Watson, J. Monetary Economics 44, 293 (1999), J. Business and Economic Statistics 20, 147 (2002), J. Am. Statistical Association 97, 1167 (2002)
M. Forni, M. Hallin, M. Lippi, L. Reichlin, Rev. Economic and Statistics 82, 540 (2000); M. Forni, M. Hallin, M. Lippi, L. Reichlin, J. Econometrics 119, 231 (2004); The Generalized Dynamic Factor Model: One-Sided Estimation and Forecastings, mimeo (1999)
J. Bai, Econometrica 71, 135 (2003)
J. Bai, S. Ng, Econometrica 70, 191 (2002)
B. Bernanke, J. Boivin, J. Monetary Economics 50, 525 (2003)
C.A. Sims, Econometrica 48, 1 (1980)
See, e.g. M. Woodford, Econometrica 58, 277 (1990); see also M. Wyart, J.P. Bouchaud, to appear in JEBO (2005)
for a recent review, see: A. Tulino, S. Verdù, Foundations and Trends in Communication and Information Theory 1, 1 (2004)
A. Edelman, N. Raj Rao, Acta Numerica, 1 (2005)
V.A. Marčenko, L.A. Pastur Math. USSR-Sb, 1, 457 (1967)
J.W. Silverstein, Z.D. Bai, J. Multivariate Analysis 54, 175 (1995)
A.N. Sengupta, P. Mitra, Phys. Rev. E 80, 3389 (1999)
L. Laloux, P. Cizeau, J.-P. Bouchaud, M. Potters, Phys. Rev. Lett. 83, 1467 (1999); L. Laloux, P. Cizeau, J.-P. Bouchaud, M. Potters, Int. J. Theor. Appl. Finance 3, 391 (2000); M. Potters, J.-P. Bouchaud, L. Laloux, Acta Physica Polonica B 36, 2767 (2005)
Z. Burda, A. Görlich, A. Jarosz, J. Jurkiewicz, Phys. A 343, 295 (2004)
See the proceedings of the conference on Applications of Random Matrix Theory, published in Acta Physica Polonica B 36, 2603 (2005)
G. Kapetanios, A new Method for Determining the Number of Factors in Factor Models with Large Datasets, mimeo Univ. of London (2004)
See, e.g., W.H. Press, B.P. Flannery, S.A. Teukolsky, W.T. Vetterling, Numerical Recipes in C : The Art of Scientific Computing (Cambridge University Press, 1992)
J. Baik, G. Ben Arous, S. Peche, Phase transition of the largest eigenv alue for non-null complex sample covariance matrices, http://xxx.lanl.gov/abs/math.PR/0403022, to appear in Ann. Prob.
N. El Karoui, Acta Physica Polonica B 36, 2681 (2005)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Bouchaud, JP., Laloux, L., Miceli, M. et al. Large dimension forecasting models and random singular value spectra. Eur. Phys. J. B 55, 201–207 (2007). https://doi.org/10.1140/epjb/e2006-00204-0
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1140/epjb/e2006-00204-0