werleycordeiro / Dynamic_Nelson_Siegel_Svensson_Kalman_Filter

Python Package: Fitting and Forecasting the yield curve
https://pypi.org/project/Dynamic-Nelson-Siegel-Svensson-Kalman-Filter/0.1.0/
MIT License
33 stars 6 forks source link

Initial parameters #1

Open mtomic123 opened 2 years ago

mtomic123 commented 2 years ago

How did you choose the initial parameters for the scipy minimization? I am looking to calibrate the Dynamic NSS for my own data using optimize.minimize, however it does not converge as I used your initial parameters which might not be suited for my data.

werleycordeiro commented 2 years ago

You can estimate the initial parameters for your data by OLS in the two-step approach using the files in the following link:

https://github.com/werleycordeiro/Dynamic-Nelson-Siegel