0

I'm wondering if there is a standard framework or methodology to decompose government bond yields across various maturities (e.g., short-, medium-, and long-term). Specifically, I’m interested in understanding what the main components are in terms of the yield for bonds of different maturities.

Are there any widely accepted models or approaches used in the finance fields to break down these yields? If so, how do they decompose government bond yields, and does this vary by maturity?

StatsScared
  • 634
  • 5
  • 12

1 Answers1

1

I share @d-stanley 's sentiment about the ambiguity of your question, but perhaps I can give you a few places to start.

There's an enormous amount of academic research on the term structure. One "decomposition" (compression is more accurate) is the principal components analysis performed by Litterman and Scheinkman in their 1991 article, "Common Factors Affecting Bond Returns," published in the Journal of Fixed Income. Loosely speaking, they show that the yield curve moves around primarily for three reasons.

  1. Shocks that affect the entire curve moving all yields in the same direction (level factor)
  2. Shocks that affect the short- and long-ends of the curve in opposite directions (slope factor)
  3. Shocks that affect the short- and long-ends of the curve in the same direction but the middle of the curve in an opposite direction (curvature factor)

These factors are responsible for over 99% of the variation in yields.

There is also a large literature on affine term structure modeling, which "compresses" the term structure down to a few stochastic factors. Monica Piazessi has a 2010 review article on "Affine Term Structure Models" in one of the Elsevier Handbooks that is a good place to start.

Finally, you might explore the literature testing the expectations hypothesis. Much of this literature examines whether forward rates have predictive power for future interest rates - a type of decomposition.

MRR
  • 191
  • 7