No Cover Image

Working paper 630 views

The Topology of Time Series: Improving Recession Forecasting from Yield Spreads

Pawel Dlotko, Simon Rudkin Orcid Logo

EconPapers

Swansea University Author: Simon Rudkin Orcid Logo

Abstract

Recession forecasting ranges from the simplistic inference from witnessing an inverted yield curve to sophisticated models drawing data from across the macroeconomic and financial spectra. Each has advantages, in simplicity and informativeness respectively, but each suffers for these. Demonstrating...

Full description

Published in: EconPapers
Published: 2019
Online Access: https://econpapers.repec.org/RePEc:swn:wpaper:2019-02
URI: https://cronfa.swan.ac.uk/Record/cronfa51888
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract: Recession forecasting ranges from the simplistic inference from witnessing an inverted yield curve to sophisticated models drawing data from across the macroeconomic and financial spectra. Each has advantages, in simplicity and informativeness respectively, but each suffers for these. Demonstrating how the properties of yield spread time series can foretell of impending recessions we introduce data topology to economics. Through an exploration of the topology of time series we highlight an untapped source of information with the potential to significantly improve understanding of the economy. Advantageously we do so without risking the overfitting of introducing other variables.
Item Description: Working paper available at https://econpapers.repec.org/paper/swnwpaper/2019-02.htm
Keywords: Economic Cycles, Topological Data Analysis, Periodicity, Yield Curve, Persistent Homology
College: Faculty of Humanities and Social Sciences