Follow us
Publications Commentary Research People Events News Resources and Videos About IFS
Home Publications Independent nonlinear component analysis

Independent nonlinear component analysis

Cemmap Working Paper CWP46/19

The idea of summarizing the information contained in a large number of variables by a small number of “factors” or “principal components” has been broadly adopted in economics and statistics. This paper introduces a generalization of the widely used principal component analysis (PCA) to nonlinear settings, thus providing a new tool for dimension reduction and exploratory data analysis or representation. The distinguishing features of the method include (i) the ability to always deliver truly independent factors (as opposed to the merely uncorre-lated factors of PCA); (ii) the reliance on the theory of optimal transport and Brenier maps to obtain a robust and efficient computational algorithm; (iii) the use of a new multivariate additive entropy decomposition to determine the principal nonlinear components that capture most of the information content of the data and (iv) formally nesting PCA as a special case, for linear Gaussian factor models. We illustrate the method’s effectiveness in an application to the prediction of excess bond returns from a large number of macro factors.

More on this topic

Cemmap Working Paper CWP56/19
A new quantile regression model for survival data is proposed that permits a positive proportion of subjects to become unsusceptible to recurrence of disease following treatment or based on other observable characteristics.
Cemmap Working Paper CWP57/19
The classical problem of the monopolist faced with an unknown demand curve is considered in a simple stochastic setting.
Cemmap Working Paper CWP58/19
This paper examines the trends in geographic localization of knowledge spillovers via patent citations, extracting multiple cohorts of new sample US patents from the period of 1976-2015.
Cemmap Working Paper CWP59/19
From its inception, demand estimation has faced the problem of “many prices.”
Cemmap Working Paper CWP60/19
This paper applies a novel bootstrap method, the kernel block bootstrap, to quasi-maximum likelihood estimation of dynamic models with stationary strong mixing data.