MSSA vs. Multivariate Regularized Expectation Maximization for Data Cleaning

22 Pages Posted: 20 Jul 2016

See all articles by Jan Dash

Jan Dash

Fordham University; Bloomberg LP

Yan Zhang

Bloomberg LP

Date Written: July 19, 2016

Abstract

Previously we introduced Singular Spectrum Analysis SSA and its multivariate extension MSSA as a powerful tool for cleaning data. Here we compare MSSA with the data filling algorithm M-REM (Multivariate Regularized Expectation Maximization). We compare theoretical methodology, numerical stability, algorithm capability, flexibility and speed. Theory and numerical tests indicate that MSSA is superior to M-REM for data cleaning.

Keywords: Data Cleaning, SSA, MSSA, M-REM, Multivariate Regularized Expectation Maximization

JEL Classification: C1, C14, C22, C63, E44, F65, G1, Y1

Suggested Citation

Dash, Jan and Zhang, Yan, MSSA vs. Multivariate Regularized Expectation Maximization for Data Cleaning (July 19, 2016). Available at SSRN: https://ssrn.com/abstract=2811717 or http://dx.doi.org/10.2139/ssrn.2811717

Jan Dash (Contact Author)

Fordham University ( email )

113 W. 60th St
New York, NY 10023
United States

Bloomberg LP ( email )

731 Lexington Ave
New York, NY 10022
United States

Yan Zhang

Bloomberg LP ( email )

731 Lexington Ave
New York, NY 10022
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
51
Abstract Views
543
rank
471,315
PlumX Metrics