We present a constructive identification proof of p-linear decompositions of q-way arrays. The analysis is based on the joint spectral decomposition of a set of matrices. It has applications in the analysis of a variety of latent-structure models, such as q-variate mixtures of p distributions. As such, our results provide a constructive alternative to Allman, Matias and Rhodes [2009]. The identification argument suggests a joint approximate-diagonalization estimator that is easy to implement and whose asymptotic properties we derive. We illustrate the usefulness of our approach by applying it to nonparametrically estimate multivariate finite-mixture models and hidden Markov models.
Authors
Research Fellow Sciences Po and University College London
Jean-Marc is a Research Fellow of the IFS and a Professor of Economics at Sciences Po, Paris, and University College London.
University of Cambridge
Professor of Economics University of Chicago
Working Paper details
- DOI
- 10.1920/wp.cem.2014.1814
- Publisher
- Institute for Fiscal Studies
Suggested citation
S, Bonhomme and K, Jochmans and J, Robin. (2014). Nonparametric spectral-based estimation of latent structures. London: Institute for Fiscal Studies. Available at: https://ifs.org.uk/publications/nonparametric-spectral-based-estimation-latent-structures (accessed: 19 April 2024).
Related documents
More from IFS
Understand this issue
Where next for the state pension?
13 December 2023
Social mobility and wealth
12 December 2023
Autumn Statement 2023: IFS analysis
23 November 2023
Policy analysis
Recent trends in and the outlook for health-related benefits
19 April 2024
Progression of nurses within the NHS
12 April 2024
Regional variation in earnings and the retention of NHS staff in Agenda for Change bands 1 to 4
10 April 2024
Academic research
Labour market inequality and the changing life cycle profile of male and female wages
15 April 2024
Interpreting cohort profiles of lifecycle earnings volatility
15 April 2024
Parental beliefs, perceived health risks, and time investment in children
15 April 2024