In the present academic work we implement the Nelson and Siegel Segmented Model (2017) in order to predict the structure of interest rates. On the other hand we compare the performance of the Segmented model with the Nelson and Siegel Classic model. The present work was done with daily data Colombian yields between the years 2013 and 2016, the data was obtained thanks to Precia who is the price provider for valuation in Colombia. We find that in the Segmented Model by locally adjusting the segments provides substantial improvements inside and outside the sample in comparison with the Classic model.
