When the yield curve is modelled using an affine factor model, residuals may still contain relevant information and do not adhere to the familiar white noise assumption.This paper proposes a pragmatic way to improve out of sample performance for yield curve forecasting. The proposed adjustment...

The appearance of negative bond yields presents significant challenges for the fixed income markets, which mainly concern related forecasting models. The Nelson-Siegel-Svensson model (NSS) is one of the models that is most frequently used by central banks to estimate the term structure of...

Issues like structural breaks and misspecification biases make it difficult to find a term structure of interest rates forecast model that dominates all competitors. Focusing on Brazilian data, this paper aims to identify the existence of combining methods that provide superior performance than...

We generalize the arbitrage-free Nelson Siegel (AFNS) model to allow λt to vary over time. We find that the time-varying λt, which determines the relative factor loadings, typically reaches its local peak before starting to decline right before a recession. Through conducting extensive...

This study examines the pricing of municipal bonds before and after a currency shock in Switzerland. Two approaches are used to decompose the municipal to treasuries bond spreads into liquidity, maturity, and default risk premiums. The first approach is the model of the cross-sectional...

In this paper we model and predict the term structure of US interest rates in a data-rich and unstable environment. The dynamic Nelson-Siegel factor model is extended to allow the model dimension and the parameters to change over time, in order to account for both model uncertainty and sudden...

This paper constructs a 10-year realized term premium from the 10-year zero coupon Treasury yield in year 0 and the ex post 3-month Treasury yields from years 0 to 10. The realized term premium swung wildly until the mid-1980's, and then fluctuated within a fairly stable range showing no trend....

When the yield curve is modelled using an a ffine factor model, residuals may still contain relevant information and do not adhere to the familiar white noise assumption. This paper proposes a pragmatic way to improve out of sample performance for yield curve forecasting. The proposed adjustment...

We propose a highly optimized latent factor representation of the yield curve obtained by training a variational autoencoder (VAE) to curve data from multiple currencies. A curious byproduct of such training is a "world map of latent space" where neighbors have similar curve shapes, and distant...