A QR Decomposition Approach to Factor Modelling: A Thesis Report

11/27/2018
by   Immanuel Manohar, et al.
0

An observed K-dimensional series { y_n} _n=1^N is expressed in terms of a lower p-dimensional latent series called factors f_n and random noise ε_n. The equation, y_n=Qf_n+ε_n is taken to relate the factors with the observation. The goal is to determine the dimension of the factors, p, the factor loading matrix, Q, and the factors f_n. Here, it is assumed that the noise co-variance is positive definite and allowed to be correlated with the factors. An augmented matrix, M̃[< a r r a y >] is formed using the observed sample autocovariances Σ̃_yy(l)=1/N-l∑_n=1^N-l(y_n+l-y̅)(y_n-y̅)^, y̅=1/N∑_n=1^Ny_n. Estimating p is equated to determining the numerical rank of M̃. Using Rank Revealing QR (RRQR) decomposition, a model order detection scheme is proposed for determining the numerical rank and for estimating the loading matrix Q. The rate of convergence of the estimates, as K and N tends to infinity, is derived and compared with that of the existing Eigen Value Decomposition based approach. Two applications of this algorithm, i) The problem of extracting signals from their noisy mixtures and ii) modelling of the S&P index are presented.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
09/02/2023

Robust estimation for number of factors in high dimensional factor modeling via Spearman correlation matrix

Determining the number of factors in high-dimensional factor modeling is...
research
06/24/2023

Latent Factor Analysis in Short Panels

We develop inferential tools for latent factor analysis in short panels....
research
02/22/2020

Block-Term Tensor Decomposition: Model Selection and Computation

The so-called block-term decomposition (BTD) tensor model has been recen...
research
12/31/2021

Modelling matrix time series via a tensor CP-decomposition

We propose to model matrix time series based on a tensor CP-decompositio...
research
09/29/2022

Modeling High-Dimensional Matrix-Variate Observations by Tensor Factorization

In the era of big data, it is prevailing of high-dimensional matrix-vari...
research
06/20/2021

Online Rank-Revealing Block-Term Tensor Decomposition

The so-called block-term decomposition (BTD) tensor model, especially in...
research
05/30/2022

Bayesian Low-Rank Interpolative Decomposition for Complex Datasets

In this paper, we introduce a probabilistic model for learning interpola...

Please sign up or login with your details

Forgot password? Click here to reset