Forecasting the Term Structure of Interest Rates with Dynamic Constrained Smoothing B-Splines
Round 1
Reviewer 1 Report
This is a nice paper. Nicely executed and clearly explained.
I only have one important suggestion to make: when the authors present their results, their description sounds a little too streamlined. I would be nice to read a more substantial discussion of the reported empirical outcomes.
A more thorough interpretation of the results is probably necessary. For example, some of the time-varying coefficients are found to exhibit unit roots and to be cointegrated. What does it mean, in economic terms? Or better, why does it matter and what are the implications, especially to investors and policy makers?
Author Response
Dear Reviewer,
Indeed, the economic interpretation of the cointegration was needed. We added the two following paragraphs on this topic on the Results section.
In economic terms, the cointegration of AFNS coefficients describes a strong relationship between long and medium term contracts, which can be a result of a political measure or some market characteristic that stimulated the emission of long term contracts based on the price of medium term contracts and vice-versa.
Since B-Splines coefficients have a more local specific behavior, these cointegrations give a more detailed analysis than AFNS model. It reveals the binding between short and medium term contracts on one side and medium and long term contracts on the other side. Likewise the economic interpretation of AFNS model, this is an important feature of the model, because it shows to investors and policy makers the magnitude of how the supply and demand on a type of contract can influence the price of other type of contract.
Thank you very much for your time and very good suggestions.
Reviewer 2 Report
The paper is well written, documented and the conclusion are supported by the methodology used. Just one minor comment: in order to insure reproducibility of the results, I if it's possible to upload the codes used to analyze the results.
Author Response
Dear Reviewer,
Thank you very much for your time. Please, download the source code produced during the research on the following link (we suggest you to use an anonymous tab on your browser). Unfortunantely it is not possible to upload the source code as an attached file here.
https://drive.google.com/open?id=1eFxUfcIMVInL3H2e_jkSyeXjRL_sxZjg
There are two folders in the zip file:
- 1. Curve models: a Qt/C++ toolkit to generate curves and export the coefficients
- 2. Analysis: an R code for time series analysis and plots generation.
Reviewer 3 Report
In my opinion, this work is complete. I would just like to see a greater discussion in the conclusion about the applications of DCOBS in future work, models (theory and applications). Maybe the title does not need the abbreviation (removing "DCOBS:")
Author Response
Dear Reviewer,
We added two paragraphs on the Results section discussing the economic implications of the cointegration of coefficients, which is a rich feature of the model and can be used to detect the influence of of a type of contract in another type of contract. Also, we removed the DCOBS from the title.
Thank you very much for your time and great suggestions.
