Taking the Cochrane-Piazzesi Term Structure Model Out of Sample: More Data, Additional Currencies, and FX Implications

Abstract

We examine the Cochrane and Piazzesi (2005, 2008) model in several out-of-sample analyzes. The model's one-factor forecasting structure characterizes the term structures of additional currencies in samples ending in 2003. In post-2003 data one-factor structures again characterize each currency's term structure, but we reject equality of the coefficients across the two samples. We derive some implications of the model for the predictability of cross-currency investments, but we find little support for these predictions in either pre-2004 or post-2003 data. The model fails to beat historical average returns in recursive out-or-sample forecasting of excess rates of return for bonds and currencies.

Monetary Policy and Exchange Rate Returns: Time-Varying Risk Regimes

“…Although some coefficient values are consistent across time (e.g., the three risk measures -vol1m, vol2m and VIX -consistently have a positive sign when they are significant), the instability of some of the model parameters across the two subperiods is striking in both Tables 5 and 6. Consistent with the findings reported in Hodrick and Tomunen (2018), for DMs, the coefficient on carry switches from positive significant to negative insignificant (in our out-ofsample results, we will show that it is negative and significant for the early part of the second subperiod). We return to focus on understanding this reversal in Section 5.…”

“…We also consider the empirical puzzle of the recent reversal of the role of the interest rate differential as a positive predictor of the exchange rate, as documented in Hodrick and Tomunen (2018). We find that this reversal reflects changes in the risk environment over time.…”

Reconstructing the yield curve

“…The main driver between their tent shape and our M shape is multicollinearity: f t (2), f t (3), f t (4) are highly correlated with correlations between 0.97 and 0.98 in both ours andFama and Bliss (1987)'s data. The fact that multicollinearity drives the tent shape away is also found byHodrick and Tomunen (2018) in their international study. Moreover, it also changes the tent shape by simply extending theFama and Bliss (1987) data (see the middle panel of Figure B.1).…”

A robust model for the term structure of interest rates: some applications in Colombia

“…After the mathematical formalization of Gaussian Models with an Affine Structure 7 of the Yield Curve 8 (GATSM) conducted by Duffie and Kan (1996), Dai and Singleton (2000), and Duffee (2002), among other authors, the GATSM has been positioned as a standard method to understand the dynamics of the sovereign debt yield curve as a function of a reduced set of factors, with multiple empirical applications in the international literature 9 (Ang and Piazzesi, 2003;Diebold, Piazzesi and Rudebusch, 2005;Duffee, 2006;Diez de los Rios, 2009;Cochrane and Piazzesi, 2009;Rudebush and Wu, 2008;Hodrick and Tomunem, 2018).…”