Network-Constrained Covariate Coefficient and Connection Sign Estimation

25 Pages Posted: 4 Feb 2020 Last revised: 5 Feb 2020

See all articles by Matthias Weber

Matthias Weber

University of St. Gallen - School of Finance

Jonas Striaukas

UC Louvain and F.R.S.-FNRS; Louvain Finance

Martin Schumacher

University of Freiburg - Medical Center

Harald Binder

Johannes Gutenberg University Mainz - University Medical Center

Multiple version iconThere are 2 versions of this paper

Date Written: January 21, 2020

Abstract

Often, variables are linked to each other via a network. When such a network structure is known, this knowledge can be incorporated into regularized regression settings via a network penalty term. However, when the type of interaction via the network is unknown (that is, whether connections are of an activating or a repressing type), the connection signs have to be estimated simultaneously with the covariate coefficients. This can be done with an algorithm iterating a connection sign estimation step and a covariate coefficient estimation step. We develop such an algorithm and show detailed simulation results and an application forecasting event times. The algorithm performs well in a variety of settings. We also briefly describe the R-package that we developed for this purpose, which is publicly available.

Keywords: network regression, network penalty, connection sign estimation, regularized

JEL Classification: C13, C52, C53, C55

Suggested Citation

Weber, Matthias and Striaukas, Jonas and Schumacher, Martin and Binder, Harald, Network-Constrained Covariate Coefficient and Connection Sign Estimation (January 21, 2020). University of St.Gallen, School of Finance Research Paper No. . 2020/01 , Available at SSRN: https://ssrn.com/abstract=3530820 or http://dx.doi.org/10.2139/ssrn.3530820

Matthias Weber (Contact Author)

University of St. Gallen - School of Finance ( email )

Unterer Graben 21
St.Gallen, CH-9000
Switzerland

Jonas Striaukas

UC Louvain and F.R.S.-FNRS ( email )

34 Voie du Roman Pays
B-1348 Louvain-la-Neuve
Louvain la Neuve, 1348
Belgium

HOME PAGE: http://sites.google.com/site/striaukasj/

Louvain Finance ( email )

34 Voie du Roman Pays
B-1348 Louvain-la-Neuve, b-1348
Belgium
+3210479429 (Phone)
+3210479429 (Fax)

Martin Schumacher

University of Freiburg - Medical Center ( email )

Hugstetter Stra├če 49
Freiburg, 79106
Germany

Harald Binder

Johannes Gutenberg University Mainz - University Medical Center ( email )

Langenbeckstrasse 1
International Office Building 301
Mainz, D-55099
Germany

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
26
Abstract Views
535
PlumX Metrics