Polish sovereign term-premium decomposition
Adrian-Crump-Moench (ACM) affine term-structure model and the Bauer-Rudebusch-Wu small-sample bias correction (BRW), estimated weekly on the LW-NSS zero-coupon panel. Includes 10y nominal-yield decomposition and cross-country term-spread comparison.
10y nominal yield decomposition (ACM)
Decomposition of the fitted 10y zero-coupon yield into the expected average short-rate path (rf, dark blue) and the residual term premium (salmon shading). Adrian-Crump-Moench affine model on five Nelson-Siegel principal components.
10y nominal yield decomposition (BRW)
Same decomposition with the Bauer-Rudebusch-Wu indirect-inference bias correction. The BRW correction tilts mass from the rf path to the term-premium component, which matters most when the autoregressive matrix is close to a unit root.
PL term premia by horizon — ACM versus BRW
All four horizons (1y, 2y, 5y, 10y) for both models. Use the pills below to switch which horizons are visible.
Cross-country 10y-2y term spread
Polish 10y-2y zero-coupon spread compared with US (GSW Treasury) and euro-area AAA. Monthly, from the same NSS-fitted curves on the curves page.
10y term premium — PL benchmark
Headline 10y term-premium series with both ACM and BRW. The BRW correction averages roughly 2 bp above ACM on the Polish sample over 2005-2026 (smaller than the ~50 bp gap in US data because the Polish Φ matrix is less persistent).
Methodology: ACM five-factor affine term-structure model, OLS regression of monthly excess returns on lagged Nelson-Siegel principal components, no cross-sectional restrictions imposed. BRW correction follows Bauer, Rudebusch and Wu (2012, 2014), iterated bootstrap with common-random-numbers and Kilian shrinkage on the risky-recursion VAR. Estimation runs on the weekly NSS panel from the curves page. See about page for full references.