Nonlinear Time Series Analysis of Daily Stock-Bond Return Relation and Stock Market Uncertainty

Show full item record



Permalink

http://urn.fi/URN:NBN:fi:hulib-201703272374
Title: Nonlinear Time Series Analysis of Daily Stock-Bond Return Relation and Stock Market Uncertainty
Author: Hasu, Justus
Other contributor: Helsingin yliopisto, Valtiotieteellinen tiedekunta, Politiikan ja talouden tutkimuksen laitos
University of Helsinki, Faculty of Social Sciences, Department of Political and Economic Studies
Helsingfors universitet, Statsvetenskapliga fakulteten, Institutionen för politik och ekonomi
Publisher: Helsingfors universitet
Date: 2013
Language: eng
URI: http://urn.fi/URN:NBN:fi:hulib-201703272374
http://hdl.handle.net/10138/39536
Thesis level: master's thesis
Discipline: Economics
Taloustiede
Ekonomi
Abstract: For an active investor it is important to know cross-asset correlation dynamics in order to invest efficiently by using a multi-asset portfolio. Stocks and interest rates are two main asset classes and their relation has witnessed a positive development in the past decade indicating a negative return relation between stocks and government bonds. Stock and bond return relationship is well studied in the past. Long-term relationship is found to be modestly positive, but it seems that something has changed in this relationship in the late 1990s. The assumption is that the stock-bond relation is dominated by investment behavior in which the uncertainty causes rapid movement of funds from riskier assets to safe assets. Further, this investment behavior causes large variations in short-term return relation, but has also affected the long-term relation of the two in the past decade. Combining the daily stock-bond return data and uncertainty measurement it can be shown that depending on the stock market uncertainty factor the return relation varies from positive (low uncertainty) to negative (high uncertainty) on the average. This relationship can be well presented with two-state regime switching logistic smooth transition regression model, where the stock market uncertainty is used as an observable variable defining the regime switching. Results are supporting previous studies and practical experience. The long-term return relation seems to be shifted from positive to negative and the short-term variation is associated with stock market uncertainty. Two-state regime switching smooth transition regression function results also supports the long-term change especially when the data is modeled in two separate sub periods.


Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show full item record