Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/282886 
Authors: 
Year of Publication: 
2023
Series/Report no.: 
Tinbergen Institute Discussion Paper No. TI 2023-073/III
Publisher: 
Tinbergen Institute, Amsterdam and Rotterdam
Abstract: 
The relationship between excess returns and the dividend price ratio is known to be unstable. However, there is no consensus on the type of instability, i.e. few or many breaks. Differences in parameter instability affect the long-term investor in particular, as misspecification errors are exacerbated as the investment horizon increases. Therefore, we investigate the consequences of different types of break processes for a long-term investor. The break process is inferred with a mixture innovation model using Bayesian methods. This allows us to estimate the break risk and the uncertainty around it. The estimated parameters show substantial instability, with an average break probability of 20.6%. Assuming constant parameters can lead to losses of up to 16.3% in certainty equivalent return for the long-term investor, even if the break probability is small in reality. The costs of ignoring uncertainty regarding the instability are smaller, but non-negligible.
Subjects: 
Return predictability
parameter instability
mixture innovation model
long-term investing
Bayesian modeling
JEL: 
C11
C32
G11
Document Type: 
Working Paper

Files in This Item:
File
Size
1.39 MB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.