I am grateful to Karim Abadir, Hans Dewachter, John Hutton, Peter Phillips, Mark Salmon, Peter Smith, Andy Tremayne, Mike Wickens, Zhuoshi Liu, and participants in the Econometric Society European Meetings 2006 for useful comments. I also thank the referees of this journal for their many helpful suggestions.

# Stochastic Volatility in a Macro-Finance Model of the U.S. Term Structure of Interest Rates 1961–2004

Article first published online: 12 AUG 2008

DOI: 10.1111/j.1538-4616.2008.00153.x

© 2008 The Ohio State University

Additional Information

#### How to Cite

SPENCER, P. D. (2008), Stochastic Volatility in a Macro-Finance Model of the U.S. Term Structure of Interest Rates 1961–2004. Journal of Money, Credit and Banking, 40: 1177–1215. doi: 10.1111/j.1538-4616.2008.00153.x

#### Publication History

- Issue published online: 12 AUG 2008
- Article first published online: 12 AUG 2008
- Received November 2, 2006; and accepted in revised form January 23, 2008.

- Abstract
- Article
- References
- Cited By

### Keywords:

- C13;
- C32;
- E30;
- E43;
- E52

- affine term structure model;
- macro finance;
- unit root;
- stochastic volatility

### Abstract

- Top of page
- Abstract
- 1. THE GENERAL AFFINE MACRO-MODEL FRAMEWORK
- 2. THE GENERAL AFFINE YIELD CURVE FRAMEWORK
- 3. THE EMPIRICAL MODELS
- 4. CONCLUSION
- Appendices
- LITERATURE CITED

This paper generalizes the standard homoscedastic macro-finance model by allowing for stochastic volatility, using the “square root” specification of the mainstream finance literature. Empirically, this specification dominates the standard model because it is consistent with the square root volatility found in macroeconomic time series. Thus it establishes an important connection between the stochastic volatility of the mainstream finance model and macro-economic volatility of the Okun–Friedman type. This research opens the way to a richer specification of both macro-economic and term structure models, incorporating the best features of both macro-finance and mainstream finance models.

This paper develops a general affine macro-finance model of the U.S. macroeconomy and the Treasury bond market. As the name suggests the macro-finance approach allows bond yields to reflect macro-economic variables as well as latent variables representing financial market factors. It is based on the central bank model (CBM) developed by Svensson (1999), Rudebusch (2002), Smets (1999), Kozicki and Tinsley (2005), and others, which represents the behavior of the macroeconomy in terms of the output gap (*g*_{t}), inflation (π_{t}) and the short-term interest rate (*r*_{t}). The model developed in this paper allows bond yields to reflect changes in macro-economic volatility related to the underlying rate of inflation.

In turn, the behavior of bond yields helps inform the specification of the macroeconomy, yielding new insights into the operation of monetary policy. In particular, early macro-finance studies showed that although macro-economic variables provide a good description of the behavior of short rates they do not provide an adequate description of long-term yields (Kozicki and Tinsley 2001, Ang and Piazzesi 2003). This finding has spawned an important macro-econometric literature that augments the CBM with latent variables, capturing exogenous changes in inflation and interest rates (Kozicki and Tinsley 2005 provide a useful summary). This literature shows that these rates are characterized by a nonstationary common trend (or unit root) that seems to be explained by the underlying rate of inflation. It follows the standard macro-econometric literature in assuming a homoskedastic (fixed) variance structure. This situation is familiar to macro-economic modelers but poses a potential problem for term structure researchers: it is well known that asymptotic (long-maturity) yields are not properly defined if the interest rate is driven by a random walk (a homoskedastic unit root process).

This theoretical problem was first raised as an empirical issue by Dewachter and Lyrio (2006), but with this notable exception, macro-finance modelers have avoided it by assuming that the underlying inflation variable follows a near-unit-root process (Ang and Piazzesi 2003, Rudebusch and Wu 2003, Dewachter, Lyrio, and Maes 2006). However, because this variable is stationary, it mean-reverts to a constant rather than the variable end-point suggested by unit root macro-economic models. As Kozicki and Tinsley (2005), Dewachter and Lyrio (2006), and others note, this means that it cannot be interpreted as a long-run inflation expectation because it is anchored to a constant that cannot be influenced by monetary policy.

Mainstream finance yield curve research avoids these problems by using heteroskedastic (stochastic volatility) interest rate models based on Cox, Ingersoll, and Ross (1985, hereafter CIR). I modify their continuous time specification for use with discrete-time macro-economic data, getting a sensible forward rate asymptote without placing constraints on the roots of system. The stochastic trend is estimated using the extended Kalman filter, which is also standard in the mainstream finance literature. Model restrictions allow the stochastic trend-volatility term to be interpreted as an inflation trend, consistent with the hypothesis that macro-economic volatility is influenced by the underlying rate of inflation (Okun 1971, Friedman 1977, Engel 1982).^{1} This specification encompasses the standard macro-finance model, which is decisively rejected by the data.

The research described in this paper was initially motivated by my interest in the asymptotic yield problem raised by early drafts of the Dewachter and Lyrio (2006) paper. I expected the new specification to outperform the standard one in explaining long maturity yields and with this in mind I extended the conventional (10-year maximum) maturity data set to include a 15-year yield, the longest available historically. However, the results are surprising in this respect. The new specification does give a dramatic improvement in fit, but the main reason for this is the importance of the Okun–Friedman heteroskedasticity effect found in the macro-data. Once this is allowed for, its more flexible yield curve specification adds very little. This finding suggests that this new macro-finance framework—which uses estimates of macro-economic volatility to inform the stochastic volatility parameters of the term structure model—should be better at discriminating between rival models than the mainstream finance one (Chen and Scott 1993, Dai and Singleton 2002), which does not. It further suggests that CBM-based studies of monetary policy should use the heteroskedastic framework rather than the current homoskedastic one, allowing the effects of the stochastic trend on the second as well as the first moments of the system to be analyzed.

The paper is set out as follows. The next section describes the macro-economic model and its stochastic structure, supported by Appendix A. Section 2 derives the bond pricing model, supported by Appendix B. It discusses the theoretical problems posed by the unit root in the standard specification and shows how these are avoided in the general affine specification. The two respective empirical models are compared in Section 3. Section 4 offers a brief conclusion and suggestions for further research.

### 1. THE GENERAL AFFINE MACRO-MODEL FRAMEWORK

- Top of page
- Abstract
- 1. THE GENERAL AFFINE MACRO-MODEL FRAMEWORK
- 2. THE GENERAL AFFINE YIELD CURVE FRAMEWORK
- 3. THE EMPIRICAL MODELS
- 4. CONCLUSION
- Appendices
- LITERATURE CITED

This section specifies the macro-economic framework. This is “general affine” or “exponential-affine” in the sense of Duffie and Kan (1996) and Duffie, Filipovic, and Schachtermayer (2003, to be explained). It consists of a heteroscedastic macro-economic vector autoregression (VAR) augmented by two latent variables, which is specified under the physical (or observed) probability measure . The yield model is specified under the risk neutral measure in the next section.

#### 1.1 The Macro-Economic Dynamics

The macro-model is based on the CBM. It represents the behavior of the macroeconomy in terms of the inflation rate (π_{t}), output gap (*g*_{t}) and the 3-month Treasury Bill rate (*r*_{1,t}). These are part of an *n*-vector *z*_{t} of macro-economic variables driven by the difference equation system:

- (1)

where *G* is a lower triangular matrix, η_{t} is an *n*-vector of independent and identically distributed (i.i.d.) orthogonal errors, and *y*_{t} is a *k*-vector of latent factors. These follow the first-order process:

- (2)

where ɛ_{t} is a *k*-vector of i.i.d orthogonal errors, θ={θ_{1},…, θ_{k}}′ and Ξ=*Diag*{ξ_{1},…, ξ_{k}}.^{2} It is assumed that *z*_{t} is observed without measurement error and that *y*_{t} is unobservable. I estimate *y*_{t} using the extended Kalman filter (Harvey 1989, Duffee and Stanton 2004) as described in Appendix C.

The specific model developed in this paper defines *z*_{t}={*g*_{t}, π_{t}, *r*_{1,t}}′, , ɛ_{t}=, and . In my preferred model, is a martingale driving the inflation asymptote: , where is a shift constant. The central tendency **r*** of the real interest rate **r** is plus the constant is assumed to be a mean-reverting Gaussian variable , so plays an identical role to , which it is convenient to set to zero, making the long-run mean. The central tendency of the nominal interest rate is thus , which reverts to the asymptote The output gap is also assumed to be a zero-mean-reverting variable: *g**_{t}= 0. These equilibrium conditions are enforced by imposing a set of restrictions on (1):

where

- (3)

to give the equilibrium solution *z**_{t}= (*I*−Σ^{L}_{l=1}Φ_{l})^{−1}Φ_{0}(*y*_{t}+ϕ) =*R*(*y*_{t}+ϕ).

This system can be consolidated by defining *x*_{t}={*y*′_{t}, *z*′_{t}}′; *v*_{t}={ɛ′_{t}, η′_{t}}′ and combining (1) and (2) to get an *L*th order difference system described in Appendix A as (A1). The yield model employs the state space form (Harvey 1989), obtained by arranging this as first-order difference system describing the dynamics of the state vector:

- (4)

where *X*_{t}={*y*′_{t}, *z*′_{t}, … , *z*′_{t-1}}′ is the state vector, *W*_{t}=*C*⋅{ɛ′_{t}, η′_{t}, 0_{1,N−k−n}}′, and Θ, Φ, and *C* are defined in Appendix A. *X*_{t} has dimension *N*=*k*+*nl*.

The macro-economic data were provided by Datastream and are shown in Figure 1. π_{t} is the annual CPI inflation rate and *r*_{1,t} is the 3-month Treasury Bill rate. The output gap series *g*_{t} is the quarterly OECD measure, derived from a Hodrick–Prescott filter. The yield data were taken from McCulloch and Kwon (1991), updated by the New York Federal Reserve Bank.^{3} These have been extensively used in the empirical literature on the yield curve. To represent this curve I use 1-, 2-, 3-, 5-, 7-, 10-, and 15-year maturities. These yield data are available on a monthly basis, but the macro-economic data dictated a quarterly time frame (1961Q4–2004Q1, a total of 170 periods). The quarterly yield data are shown in Figure 2. The 15-year yield is typically shown shown at the top of the figure, while the shorter maturity yields are most frequently shown at the bottom; however, reversals in this ordering tend to occur near business cycle peaks.

These inflation and interest rates all exhibit a high degree of persistence, which could be the effect of slow mean reversion, unit roots, or structural breaks. Table 1 shows the means, standard deviations, and first-order autocorrelation coefficients of these data, as well as KPSS (Kwiatowski et al. 1992) and adjusted Dickey–Fuller (ADF) test results. The ADF tests show that the null hypothesis of nonstationarity for these variables cannot be rejected at the 5% level. The KPSS statistics for inflation and 1- to 10-year interest rates are only significant at the 10% level, suggesting that the null hypothesis of stationarity may just be acceptable. However, the KPSS statistic for the 15-year rate is almost significant at the 5% level. Moreover, Fama (2006) argues persuasively that the long upswing and downswing in rates evident in the figures was the result of a succession of permanent shocks that were on balance positive until 1981 and negative thereafter. In this paper, I follow Kozicki and Tinsley (2005), Dewachter and Lyrio (2006), and Fama (2006) in analyzing a macro-econometric model characterized by a unit root.

g | π | r_{1} | r_{4} | r_{8} | r_{12} | r_{20} | r_{28} | r_{40} | r_{60} | |
---|---|---|---|---|---|---|---|---|---|---|

^{}Note: Inflation (π) and 3-month T-Bill rate ( *r*_{1}) are from Datastream. Output gap (*g*) is from OECD. Yield data are U.S. Treasury discount bond equivalent data compiled by McCulloch and Kwon (1990) updated by the New York Federal Reserve Bank. Mean denotes sample arithmetic mean expressed as percentage per annum; Std. is standard deviation and Auto. is the first-order quarterly autocorrelation coefficient. Skew. and Kurt. are standard measures of skewness (the third moment) and kurtosis (the fourth moment). KPSS is the Kwiatowski et al. (1992) statistic testing the null hypothesis of level stationarity. The 10% and 5% significance levels are 0.347 and 0.463, respectively. ADF is the adjusted Dickey–Fuller statistic testing the null hypothesis of nonstationarity. The 10% and 5% significance levels are 2.575 and 2.877, respectively.
| ||||||||||

Mean | 0.0266 | 4.4345 | 5.8051 | 6.3954 | 6.6305 | 6.7849 | 7.0021 | 7.1478 | 7.2513 | 7.4383 |

Std. | 2.3382 | 2.9841 | 2.7750 | 2.8174 | 2.7355 | 2.6508 | 2.5447 | 2.4790 | 2.4194 | 2.3833 |

Skew. | −0.575 | 1.4161 | 1.1907 | 1.8820 | 1.8697 | 1.8793 | 1.9633 | 1.9622 | 1.9215 | 1.8188 |

Kurt. | 1.0230 | 1.5178 | 2.3986 | 1.1986 | 1.0622 | 0.9505 | 0.9166 | 0.7550 | 0.5929 | 0.3945 |

Auto. | 0.4632 | 0.9921 | 0.9815 | 0.9892 | 0.9923 | 0.9944 | 0.9953 | 0.9963 | 0.9969 | 0.9971 |

KPSS | 0.2151 | 0.3399 | 0.3100 | 0.3307 | 0.3348 | 0.3399 | 0.3475 | 0.3548 | 0.3761 | 0.4298 |

ADF | −4.133 | −2.411 | −2.110 | −2.100 | −2.063 | −2.031 | −2.043 | −1.991 | −1.951 | −2.091 |

#### 1.2 The Stochastic Structure

The standard macro-finance model assumes that the volatility structure is homoskedastic and Gaussian: *W*_{t}∼*N*(0_{N}, Ω), while mainstream finance models usually assume that volatility is stochastic, driven by square root processes in one or more of the state variables.^{4}Dai and Singleton (2000) derive “admissibility” conditions to ensure that these state variables remain nonnegative and the variances are well defined. They classify an admissible model with *N* state variables and *m* independent square root factors conditioning volatility is classed as Thus the standard macro-finance model (which is homoskedastic) is classified as and the mainstream specification (with a single stochastic volatility term) as .

This paper develops a model that encompasses the and specifications. These models all generate affine yield curves because the probability distributions underpinning them are all “exponential-affine” in the sense of Duffie, Filipovic, and Schachtermayer (2003). They define a process as exponential-affine under any measure if the conditional moment generating function (MGF) ( | *X*_{t}]) for *X*_{t+1}, is an exponential-affine (log-linear) function of *X*_{t}. *E*^{M} denotes the expectation under the measure while *E* and *V* denote the mean and variance under the state price density and ν is a vector of Laplace coefficients. The MGF is the Laplace transform of the density of *X*_{t+1}. For example assumes that ɛ_{1,t} is normally distributed with mean zero and standard deviation δ_{01} and we use the formula for the MGF of a normal variable:

- (5)

In , this latent variable also drives volatility through a square root process similar to the diffusion for the spot interest rate in CIR. They show that in discrete time, this has a noncentral conditional χ^{2} distribution. If we normalize the time interval (*s*−*t*) in their equation (18) as unity and replace *r* by :

- (6)

where 2*c* is the scale factor, is the noncentrality parameter, and 2*c*μ shows the degrees of freedom.

This process is exponential-affine because its conditional MGF is a log-linear function of :

- (7)

(provided that ν < *c*,Johnson and Kotz 1970). Differentiating (7) with respect to ν once, twice, and then setting ν to zero gives the conditional mean and variance:

where

- (8)

In the limiting case of a unit root, the degree of freedom parameter is zero. This model is studied by Seigel (1979) and his basic results are reported in chapter 29 of Johnson and Kotz (1970). Important results have also been obtained for this case by Gourieroux and Jasiak (2002). In this limit: ; δ_{01}=μ= 0; *c*= 2/δ_{11}, and (8) simplifies to:

- (9)

This process is a martingale: the expectation of any future value is equal to the current value. However, unlike the random walk model, the error variance is also proportional to this value. These models can all be represented as: To be consistent with (2) we set the intercept equal to μ for the models (5) and (8) and to zero in (9).

In , this stochastic trend also conditions the volatility of the other variables. It is ordered as , the first variable in the (*k*+*n*) vector *x*_{t}. The other contemporaneous variables are put into an (*k*+*n*− 1) vector *x*_{2,t}, so that: *x*_{t}={*x*_{1,t}, *x*′_{2,t}}′ and conformably: *v*_{t}={*w*_{1,t}, *v*′_{2,t}}′ and *w*_{t}={*w*_{1,t}, *w*′_{2,t}}′, where . Similarly, writing *X*_{t}={*x*_{1,t}, *X*′_{2,t}}′ and partitioning *W*_{t}, Θ, Φ,*C* conformably (see Appendix A), (4) becomes:

- (10)

where , and . In this paper subscripts 1 and 2 denote partitions of *N* (or *k*+*n*) dimensional vectors and matrices into 1 and *N*− 1 (or *k*+*n*− 1). The stochastic structure for (10) is described in Appendix A. The distribution of *x*_{2,t} and *x*_{2,t} conditional upon *x*_{1,t−1} is assumed to be Gaussian. The conditional covariance of *x*_{2,t} is Σ_{0}+Σ_{1}*x*_{1,t−1}, where: Σ_{i}=*C*_{22}Δ_{i}*C*′_{22} and Δ_{i}=*Diag*{{δ_{i2}, … , δ_{i(k+n)}}, 0′_{N−k−n}}; *i*= 0, 1.*C*_{22} is a lower triangular and Δ_{i}; *i*= 0, 1 are deficient diagonal (*N*− 1)^{2} matrices.

This model is the discrete time analogue of the “general affine” model developed by Duffie and Kan (1996), which generalizes the CIR model by adding a translation or shift constant to a model variable like an interest rate or stochastic trend when defining the volatility term in the rate diffusion. However, it is convenient to allow for this shift by using in (3) instead, keeping a CIR process but making the inflation asymptote π* a Duffie and Kan (1996) process. Importantly, the Duffie and Kan model encompasses . Similarly, the discrete time model is a special case of my model, which can be obtained by taking the limit as *c* tends to infinity. Setting δ_{11} and Σ_{1} to zero then makes *x*_{1,t} and *x*_{2,t} homoskedastic. This also renders the distribution of Gaussian, allowing this to be defined in the model as another zero-mean-reverting latent variable (θ_{1}= 0).^{5} These models are “admissible” in the sense of Dai and Singleton (2000, 2002) in that they ensure that the variance structure remains nonnegative definite.^{6} They can be used independently of the yield specification, as for example by Kozicki and Tinsley (2005). However, my interest is to use them jointly with bond market data, using the macro-yield framework developed in the next section.

### 2. THE GENERAL AFFINE YIELD CURVE FRAMEWORK

- Top of page
- Abstract
- 1. THE GENERAL AFFINE MACRO-MODEL FRAMEWORK
- 2. THE GENERAL AFFINE YIELD CURVE FRAMEWORK
- 3. THE EMPIRICAL MODELS
- 4. CONCLUSION
- Appendices
- LITERATURE CITED

This section is supported by Appendix B and shows how exponential-affine MGFs can be used to model the yield curve consistently with the macro-models of the previous section. It is based on the fact (Duffie, Filipovic, and Schachtermayer 2003) that the MGF of a distribution that is exponential-affine under the risk neutral measure generates an exponential-affine discount bond price model:

- (11)

The natural logarithm of this price is denoted by *p*_{τ,t} and is linear in *X*_{t}. Reversing sign and dividing by maturity τ gives the discount yields: *r*_{τ,t}=−*p*_{τ,t}/τ=α_{τ}+β′_{τ}*X*_{t}, where α_{τ}=γ_{τ}/τ; β_{τ}=Ψ_{τ}/τ. The slope coefficients β_{τ} are known as “factor loadings.” Stacking the yield equations for τ= 4, 8, 12, 20, 28, 40, and 60 quarters and adding an error vector *e*_{t} gives the multivariate regression model used in Section 3 for the (7 × 1) vector *r*_{t}:

where

- (12)

and *e*_{t} is an i.i.d. error vector.^{7} This is the affine yield curve framework used in this paper. The remainder of this section discusses the measure used for asset pricing and the “essentially affine” yield models and corresponding to the macro-models and of the previous section.

#### 2.1 The Risk Neutral Measure Q

Assets are priced under the risk neutral measure, which adjusts the state probabilities in such a way that they all have the same expected return. These adjustments depend upon “price of risk” parameters that show the effect of model variables on the risk premia. For the yield model to be affine these prices must also be affine in the state variables. So, for example, the variable λ_{1,t} shows the price of risk associated with the stochastic trend, which plays an important role in this analysis:

- (13)

If this is zero, then an “asymptotic” or “end-point” portfolio that is constructed so that it is only exposed to shocks in *x*_{1,t} has a zero risk premium and is expected to earn the spot rate. If it is constant (λ_{1,t}=λ_{10}), then variations in this risk premium depend only upon variations in volatility, such as those induced by *x*_{1,t} in . This parameter plays the key role in that model. If λ_{11} is also nonzero then the trend can influence the risk premia thorough variations in the price of risk, even if volatility is fixed as in , so λ_{11} plays the key role in that model. Appendix B shows how the prices of risk associated with the other variables are adjusted, following Duffee (2002). After this modification, the models and are classified as the “essentially affine” models and , respectively, represented by the empirical models M0 and M1 in Section 3. Following Dewachter and Lyrio (2006), both models incorporate the restrictions: ξ_{1}= 1; Λ_{12}= 0_{N−1}.

As Appendix B explains, variations in λ_{1,t} are a nuisance in the model M1 and the conventional assumption is that λ_{1,t}=λ_{10} for that model. On the other hand, it is important to allow λ_{1,t} to reflect variations in *x*_{1,t} in M0, so in this case λ_{1,t}=λ_{10}+λ_{11}*x*_{1,t}. This means that M1 does not encompass M0. However, an encompassing specification that nests both models can be constructed by relaxing the usual macro-finance assumption that bond market participants use the true value of ξ_{1} in pricing. This gives my “baseline” model M2, which uses a new parameter to describe their estimate of the speed of adjustment of *x*_{1,t} under , which may differ from the parameter ξ_{1} defined in the macro-model.^{8} This modification is a technical one, designed to allow encompassing tests, but could detect an error in the market estimate of ξ_{1}. Appendix B derives the MGF of M2 under measure and shows how it can be used as a “model generating function” to derive the other empirical models. It can for instance be used as a moment generating function to give the dynamics of this system under :

- (14)

where the time *t* expectations of the error terms are zero under . Table 2 shows parameter values for models M0-M2, where

- (15)

These reduced form parameters show the effects of the coefficients λ_{10} and λ_{11} modeling the price of risk associated with *x*_{1,t}. The vectors Λ_{20} and Λ_{21} and the matrix Λ_{22} are defined in (B6) and model the price of risk associated with *x*_{2,t}. Obviously, if all of these parameters are set to zero, the parameters revert to those shown (for ) in the first column. The parameter values for model M2 are shown in the second column. Those for M1 in the next column assume the market uses the true value of ξ_{1}. The parameters for M0 in the final column are standard and Appendix B shows that they can be derived from those for M2 by setting δ_{11} and Σ_{1} to zero and taking the limit as *c* tends to infinity. In this case we replace by , which is the Duffee (2002) risk-adjusted parameter.

^{}Note: This table shows how the shift from the state probability measure underpinning the macro-model to the risk-neutral measures used to price financial assets shift the coefficients of (14). The scalars λ _{10}and λ_{11}reflect the price of risk (or risk–reward ratio) associated with variations in*x*_{1,t}, while*H*_{i}and**ϒ**_{i};*i*= 0, 1 are matrices of parameters reflecting the price of risk associated with the other state variables and are defined in (15). The baseline model M2 allows to differ from the parameter ξ_{1}defined in the macro-model. M1 is the*EA*_{1}(*N*) model with . M0 specializes M2 by setting δ_{11}and Σ_{1}to zero and taking the limit as*c*tends to infinity. This gives the standard*EA*_{0}(*N*) model with the Duffee (2002) risk adjustments shown in the final column.
| ||||

Model: | M2 | M1 | M0 | |

Type: | EA_{1}(N) | EA_{0}(N) | ||

Measure: | P | Q | Q_{1} | Q_{0} |

θ_{1} | ||||

Θ_{2} | ||||

Φ_{21} | ||||

Φ_{22} | Φ^{Q}_{22}=Φ_{22}−Λ_{22} |

* 2.2 The**EA*_{1}(*N*)*Yield Specifications (M1 and M2)*

The coefficients of (11) are partitioned Ψ_{τ} conformably with (10) as {ψ_{1,τ}, Ψ′_{2,τ}}′. They are recursive in maturity. Since −*p*_{τ,1}=*r*_{1,t} its coefficients have the starting values γ_{1}=ψ_{1,1}= 0; and Ψ_{2,1}=*J*_{2,r}, where *J*_{2,r} is a selection vector such that *J*′_{2,r}*X*_{2,t}=*r*_{1,t}. It is also recursive in the sense that Ψ_{2,τ} does not depend upon ψ_{1,τ−1} (or γ_{τ−1}):

- (16)

where (defined in Table 2) adjusts Φ_{22} to allow for the effect of *X*_{2,t} on the associated prices of risk. I assume that the roots of this subsystem are stable under , so this has the asymptote:

- (17)

Dividing by τ and taking the limit as this goes to infinity gives the limit for β_{2} shown in Table 3. This Gaussian substructure is common to all models, but the structure of the remaining coefficients depends critically upon the model specification, particularly if the system is nonstationary.

Model | EA_{0}(N) | EA_{1}(N) |
---|---|---|

^{}Note: This table shows the effect on the yield model of imposing unit root restrictions on *x*_{1,t}under . The maturity limits for its factor loading (β*_{1}) are very different in the two models. This is reflected in the behavior of the forward rate asymptote (*f**_{t}). β*_{2}=Ψ*_{2}/τ is common to both specifications where Ψ*_{2}= lim_{τ∞}Ψ_{2,τ}=.
| ||

Limit: | ||

β*_{1}= lim_{τ∞}β_{1,τ} | 0 | |

β*_{2}= lim_{τ∞}β_{2,τ} | 0_{N−1} | 0_{N−1} |

f*_{t}= lim_{τ∞}f_{τ,t} | (Θ_{2}−H_{0})′Ψ*_{2}+τβ*_{1}x_{1,t} | |

Importantly, as Dai and Singleton (2002) and others note, nonstationarity under is not a problem in the *EA*_{1} specification. Indeed their results, like previous mainstream estimates (Chen and Scott 1993) suggest that there is a root that is significantly larger than unity under . The volatility of *x*_{1,t+1} is linear in *x*_{1,t} and so the equation determining its price coefficient includes a nonlinear Jensen effect. For M2:

- (18)

For a regular solution:

- (19)

As Campbell, Lo, and Mackinlay (1997) note in a similar heteroskedastic yield curve model,^{9} the price parameter ψ*_{1} is determined by a quadratic rather than a linear equation and is well defined (with β*_{1}= 0) even if .^{10} For the intercept:

- (20)

Unit roots are not a problem in . Indeed, they simplify the model structure, giving model M1 with . Substituting μ= 0 into (20)“switches off” the logarithmic term and makes γ_{τ} and hence the asymptotic forward rate *f** independent of ψ*_{1}, as shown in Appendix B. In other words, because the volatility of *x*_{1,t} is proportional to *x*_{1,t−1}, the associated Jensen effects are found in (18) but not in (20). This also simplifies the risk premium, derived in Appendix B as (B12).

* 2.3 The Standard**EA*_{0}(*N*)*Specification (M0)*

The pricing formulae for this model are well known and Appendix B considers them as limits of these baseline M2 formulae. Taking the limit of (18) as *c* tends to zero gives a quadratic recursion. However, the restriction δ_{11}= 0“switches off” this quadratic effect, reducing (18) to a linear recursion:

- (21)

The intercepts do follow a quadratic recursion:

- (22)

with the parameters defined in Table 2. In contrast to the M1 model, the intercept exhibits the nonlinear Jensen effects in M0 and not the first slope coefficient.

If , then clearly (21) has a unit root and as Appendix B and Table 3 show, the long forward rates behave like (−)τ^{2} in the limit, reflecting the well-known asymptotic problem. In the specification of Dewachter and Lyrio (2006), *x*_{1} has a unit root under but is mean reverting under : . This provides a neat way of avoiding the asymptotic yield problem while allowing the inflation asymptote in the macro-model to be a variable end-point as in Kozicki and Tinsley (2001). However, this restriction constrains the rate at which ψ_{1,τ} grows in the recursion (21), constraining the effect of the stochastic trend on short-maturity yields. It also means that the associated factor risk premium, the excess return expected for holding the “asymptotic” portfolio (constructed so that the portfolio weights sum to: ψ_{1}= 1, Ψ′_{2}= 0_{N−1}), is negatively related to the inflation trend, as Appendix B demonstrates. Moreover, the *EA*_{1}(*N*) model relaxes this restriction and my results, reflecting those of mainstream research (e.g., Chen and Scott 1993, Dai and Singleton 2002), suggest that this root is significantly *greater* than unity under .

### 3. THE EMPIRICAL MODELS

- Top of page
- Abstract
- 1. THE GENERAL AFFINE MACRO-MODEL FRAMEWORK
- 2. THE GENERAL AFFINE YIELD CURVE FRAMEWORK
- 3. THE EMPIRICAL MODELS
- 4. CONCLUSION
- Appendices
- LITERATURE CITED

The empirical model consists of a heteroskedastic VAR describing the three macro-economic variables (4) and the associated equations describing the seven representative yields (12). It is estimated by quasi-maximum likelihood and the extended Kalman filter, which gives optimal linear estimates of the latent variables in this situation. The likelihood function is derived in Appendix C. The preliminary tests reported in Section 1.1 indicated the presence of a unit root in the macro-economic and yield data. Further (Akaike information criterion (AIC)) tests suggested a third-order lag structure for (1). With *n*= 3, *k*= 2, and *l*= 3, there are *N*= 11 state variables. Consequently this research focussed on the and specifications. The baseline model M2 uses 66 parameters^{11} and has a log-likelihood L(2) = 747.5 as shown in Table 5.

Model M | Specification* macro/finance | Restrictions (w.r.t M4) | Parameters | Log-likelihood | ||||||
---|---|---|---|---|---|---|---|---|---|---|

k(M) | k(4)-k(M) | k(3)-k(M) | k(2)-k(M) | L(M) | L(4)-L(M) | L(3)-L(M) | L(2)-L(M) | |||

^{}Notes: (*) Model specification *A*_{m}(*N*),*EA*_{m}(*N*), where*m*= number of variables conditioning volatility and*N*= number of state variables.
| ||||||||||

Standard | ||||||||||

0 | A_{0}(N)/EA_{0}(N) | 60 | 16 | 11 | 6 | 678.2 | 142.4 | 135.6 | 138.6 | |

χ^{2}_{95} | δ_{11}= 0, Δ_{1}= 0 | 26.30 | 19.68 | 12.59 | ||||||

p | ξ_{1}= 1 | 0.0 | 0.0 | 0.0 | ||||||

1 | Stochastic trend | |||||||||

A_{1}(N)/EA_{1}(N) | 63 | 13 | 3 | 747.0 | 5.40 | 1.00 | ||||

χ^{2}_{95} | 22.36 | 7.81 | ||||||||

p | 0.97 | 0.80 | ||||||||

2 | Baseline | |||||||||

A_{1}(N)/EA_{1}(N) | 66 | 10 | ||||||||

χ^{2}_{95} | 747.5 | 4.4 | ||||||||

p | 18.31 | |||||||||

0.93 | ||||||||||

3 | Hybrid | |||||||||

A_{1}(N)/EA_{0}(N) | 71 | 5 | 745.4 | 8.60 | ||||||

χ^{2}_{95} | 11.07 | |||||||||

p | 0.13 | |||||||||

4 | Encompassing | 76 | ||||||||

A_{1}(N)/EA_{1}(N) | 749.7 | |||||||||

χ^{2}_{95} | ||||||||||

p |

The model M1 specializes this by assuming a unit root and maintaining the standard identity between macro- and yield parameters under : .^{12} These 3 restrictions are easily accepted by the data: the χ^{2}(3) likelihood ratio test gives an acceptance value of *p*= 0.97. The unit root specification M0 is also nested in M2, employing six restrictions: δ_{11} and **Δ**_{1}(4) are set to zero and again ξ_{1}= 1. However, its log-likelihood of L(0) = 678.2 is much lower than for the other models and it is overwhelmingly rejected against M2. This rejection is largely due to the restriction δ_{11}= 0. The effect of this is twofold: (i) it makes the stochastic trend homoskedastic and (ii) it removes the nonlinearity from (18), reducing it to the linear recursion (21). Theoretically, these two effects are inextricably related because the parameters of the stochastic structure structure (δ_{01}, δ_{11}, Δ_{0}, Δ_{1}, and *C*_{22}) are not affected by the change of measure and the conventional macro-finance assumption is that they are the same in the macro- and yield models. However further relaxing this assumption allows (i) and (ii) to be separated.

To explore this idea, I constructed two new models: M3 and M4. The “encompassing” specification M4 nests all of the other models. It is based on M2 but uses 10 new parameters (, and ) to replace their macro-equivalents in the yield model. It has the log-likelihood value L(4) = 749.7. As Table 5 shows, model M1 is acceptable against this alternative, suggesting that the conventional macro-finance assumption is valid. Model M3 then specializes M4 by using the restrictions and . This is a hybrid in which the true macro-model is the heteroscedastic model, but the bond market mistakenly uses a best-fit specification instead of Comparing its log-likelihood L(3) with that L(0) of M0 which it nests, gives an estimate of effect (i) of heteroskedasticity in the macroeconomy. This is highly significant. Indeed, M3 is accepted against the alternative of M4 (*p*= 0.13), with a likelihood almost as high as that for M2 and my preferred model M1. Although these models are not directly comparable, this observation tells us that (ii) the increase in fit due to the use of a macro-consistent yield model, is in practice relatively small. In other words, if the yield model parameters (, and ) are estimated separately they not very well determined statistically. In practice, these parameters are determined by the macro-finance restriction, which equates them with their macro-model analogues and thus uses macro data to pin them down.

Since these L(M) values are the sum of log-likelihood values at each period (Appendix C) they can be analyzed as time series. Figure 3 shows the effect of disaggregating the differences [L(3)–L(0)] due to effect (i) of heteroskedasticity in the macroeconomy. This is very marked in the early 1980s when the stochastic trend peaks, but also noticeable when the trend is low, at the beginning and the end of the estimation period. The difference [L(4)–L(3)] due to (ii), the use of the more general yield curve specification, is much smaller and reflecting this, the residuals from the heteroscedastic macro-based specifications M1, M2, M3, and M4 are all very similar. Despite the theoretical superiority of the -based yield specification, it is hard to see any systematic improvement over maturity or over time. The impulse responses and factor loadings of these models are also similar. For that reason I now focus on the results for the preferred model, M1.

#### 3.1 The Empirical Macro-Model

At the core of this model there is a macro VAR with a steady state solution dictated by the restrictions (3). The novelty here is the introduction of the square root volatility effects implied by the CIR-based term structure model. The model is driven by a nominal factor *x*_{1,t} and a real factor *x*_{2,t}. Model estimates of these factors are shown in Figure 4, along with their 95% confidence intervals. Most of the work is done by the nominal factor, which has a unit root. Since *x*_{1,t} has a noncentral chi-square distribution, the downside variance is smaller than the upside, but this asymmetry is only apparent at the beginning of the estimation period when the underlying inflation rate is low. This variable drives the conditional heteroskedasticity in the macro-variables. Their one-quarter-ahead forecasts values and 95% confidence intervals are shown in Figure 5. The effect of heteroskedasticity is particularly pronounced in the case of the spot rate, consistent with the finding in univariate models (Chen et al. 1992, Ait-Sahalia 1996, Stanton 1997, and others). Its variance is low over the first 4 years of the estimation period, consistent with the *ex post* stability of interest rates over this period (Figure 5c). The behavior of the spot rate over the medium term is also influenced by the real factor *x*_{2,t}, as is clear from a comparison with Figure 4(b). As expected, this real interest series reveals a marked tightening of monetary policy in the late 1970s, with a very relaxed stance in the early 1990s and again post-millennium. The model attributes the ultra-low interest rates seen over the early years of the millennium to a relaxation of monetary policy, coming against a background of a low underlying inflation rate.

How firmly do these factors anchor inflation and interest rates? This question depends upon whether output, inflation, and interest rates and the real factor *x*_{2,t} are contintegrated with the nonstationary nominal factor *x*_{1,t}. This was checked by running ADF tests on the residuals of the output, inflation, and interest rate equations, which decisively reject nonstationarity (Table 6b). These variables adjust quickly and smoothly to their equilibrium values. This mean-reversion effect can be summarized in terms of the model's eigenvalues. The autoregressive coefficient associated with *x*_{1,t} is unity, but the other roots are stable and are reported in Table 7. Four pairs of roots are sinusoidal, reflecting the cyclical nature of the macro-economic data.^{13}

g | π | r_{1} | r_{4} | r_{8} | r_{12} | r_{20} | r_{28} | r_{40} | r_{60} | |
---|---|---|---|---|---|---|---|---|---|---|

^{}Notes: The first row reports the unadjusted *R*^{2}; the second reports the root mean square error (RMSE). ADF is the adjusted Dickey–Fuller statistic testing the null hypothesis of nonstationarity. The 10% and 5% significance levels are 2.575 and 2.877, respectively. Auto. is the first order quarterly autocorrelation coefficient (with*t*-value in parentheses).
| ||||||||||

R^{2} | 0.9020 | 0.9664 | 0.9099 | 0.8616 | 0.8797 | 0.8901 | 0.9158 | 0.9297 | 0.9341 | 0.9355 |

RMSE | 0.1824 | 0.5452 | 0.8321 | 1.045 | 0.9458 | 0.8563 | 0.7359 | 0.6550 | 0.6188 | 0.6034 |

ADF | −9.4633 | −6.3951 | −6.3239 | −6.6522 | −6.7983 | −7.0042 | −7.1833 | −7.2131 | −7.0550 | −6.9903 |

Auto. | −0.1218 | −0.1283 | −0.0933 | −0.0261 | 0.0049 | 0.0948 | 0.1081 | 0.0832 | 0.0153 | −0.0021 |

(1.57) | (1.59) | (1.21) | (0.34) | (0.06) | (1.23) | (1.98) | (1.08) | (0.19) | (0.02) |

M0 | M1 |
---|---|

1 | 1 |

0.92415 ± 0.08084i | 0.93103 ± 0.10677 |

0.85770 | 84000 |

0.55264 ± 0.27966i | 0.61927 |

−0.00649 ± 0.51542i | 0.51854 ± 0.18382 |

0.51152 | 0.03600 ± 0.50129 |

−0.24485 ± 0.012387 | −0.26430 ± 0.09198 |

These cyclical effects are seen more clearly in the impulse responses, which show the dynamic effects of innovations in the macro-economic variables on the system. Because these innovations are correlated empirically, we work with orthogonalized innovations using the triangular factorization defined in (A1). The orthogonalized impulse responses show the effect on the macro-economic system of increasing each of these shocks by 1 percentage point for just one period using the Wold representation of the system. This arrangement is affected by the ordering of the macro-economic variables in the vector *x*_{t}. Like Kozicki and Tinsley (2005), I adopt the standard ordering: . The first shock (υ_{1}) reflects permanent policy or expectational changes in the inflation asymptote while the second (υ_{2}) reflects structural shocks to the real interest rate. Conventionally, υ_{3} is interpreted as a positive demand shock and (υ_{4}) as a negative supply shock. Finally, (υ_{5}) represents transitory changes in monetary policy.

Figure 6 shows the results of this exercise. This gives a plausible description of the macro-economic dynamics. As in the model of Kozicki and Tinsley (2005), the use of Kalman filters to pick up the effect of unobservable expectational influences seems to solve the notorious price puzzle—the tendency (noted originally by Sims 1992) for increases in policy interest rates to anticipate inflationary developments and apparently cause inflation. The nominal filter dictates the long-run equilibrium of the macroeconomy (and its volatility). These effects are persistent, but the responses of the macro-economic variables to surprises in inflation, output, and interest rates are rapid. They are largely exponential in nature, suggesting that monetary policy has been effective in securing its objectives quickly, without significant policy reversals or cycles.

These results are reflected in Figure 7, which report the results of the analysis of variance (ANOVA) exercise. These figures show the share of the total variance attributable to the innovations at different lag lengths and are also obtained using the Wold representation of the system, as described in Cochrane (1997). They indicate the contribution each innovation would make to the volatility of each model variable if the error process was suddenly started (having been dormant previously). So, for example, we see that the output surprise η_{1,t} accounts for nearly all of the short-run volatility in output, with similar results for the responses of inflation and spot rates to their own innovations. However, the effect of other innovations builds up over time.

#### 3.2 The Empirical Yield Model

The behavior of the yield curve is dictated by the factor loadings. These are depicted in Figure 8, as a function of maturity (expressed in quarters). The first panel shows the loadings on *x*_{1,t} (broken line), and *x*_{2,t} (continuous line). The second panel shows the loadings on π (dotted line), *g* (broken line), and the spot rate (continuous line). The spot rate provides the link between the macro-economic model and the term structure. Since it *is* the 3-month yield, this variable has a unit coefficient at a maturity of one quarter and other factors have a zero loading. The spot rate loadings decline over the next few years, reflecting the adjustment of the spot rate toward *x*_{1,t} and *x*_{2,t}. The spot rate thus determines the slope of the short-term yield curve. Three- to 5-year maturity yields are strongly influenced by the behavior of the real rate factor *x*_{2,t}. The loading on this factor then fades gradually over the longer maturities, allowing this to act as a “curvature” factor. In contrast, the loading upon *x*_{1,t} moves up to unity and then increases gradually with maturity over the 2- to 15-year maturity range, so that it acts as a “level” factor. The loadings for output and inflation have a humped shape, but are relatively small.

Figure 9 shows the risk premia implied by models M0 and M1 for the 15-year yield. Although the loadings for these models are similar, the risk premia differ because they also depend upon the way that the stochastic trend affects volatility (via λ_{10} in M1) and the price of risk (λ_{11} in M0). The parameter λ_{10} is significantly negative in M1, so the stochastic trend has a positive effect on the asymptotic risk premium in that model (Appendix B). In M0, λ_{11} adopts a positive value in order to keep the model dynamics stable under , so the stochastic trend has a negative effect on the asymptotic premium. This means that although the stochastic trend has a positive effect on the 15-year premia in both models, this is more powerful in M1 than it is in M0. The real factor also has a strong positive effect in both models, as is clear from the figure. The effect of the macro-economic variables on the risk premia in these models is relatively small in the 15-year area. The impulse responses for the 5-year yield and ANOVA results for the 10-year yield are shown in Figures 6 and 7. These reflect the combined effect of the factor loadings and the dynamic characteristics of the model variables discussed in the previous section. The behavior of the 5-year yield depends upon the spot rate and the financial factors. The variance of the 10-year yield (Figure 7) is dominated by the shocks to the two financial factors, reflecting the “level” and “curvature” effects. The effect of the spot rate and other macro-economic variables is negligible at this maturity. Table 6 shows that the joint macro-yield model closely replicates the first three moments of the data shown in Table 1.

g | π | r_{1} | r_{4} | r_{8} | r_{12} | r_{20} | r_{28} | r_{40} | r_{60} | |
---|---|---|---|---|---|---|---|---|---|---|

^{}Notes: Mean denotes sample arithmetic mean expressed as percentage per annum, and Std. is standard deviation. Skew. and Kurt. are standard measures of skewness and kurtosis.
| ||||||||||

Mean | 0.1433 | 4.4933 | 5.9103 | 6.5804 | 6.7846 | 6.9040 | 7.1637 | 7.2676 | 7.3626 | 7.4621 |

Std. | 2.23992 | 2.96308 | 2.5555 | 2.6222 | 2.5782 | 2.5277 | 2.4582 | 2.4204 | 2.3889 | 2.3193 |

Skew. | −0.553 | 1.3980 | 0.9157 | 0.7411 | 0.8196 | 0.8749 | 0.9212 | 0.9298 | 0.9238 | 0.8994 |

Kurt. | 3.9787 | 4.5119 | 4.4288 | 3.8537 | 3.8352 | 3.8021 | 3.7220 | 3.6512 | 3.5751 | 3.4752 |

### 4. CONCLUSION

- Top of page
- Abstract
- 1. THE GENERAL AFFINE MACRO-MODEL FRAMEWORK
- 2. THE GENERAL AFFINE YIELD CURVE FRAMEWORK
- 3. THE EMPIRICAL MODELS
- 4. CONCLUSION
- Appendices
- LITERATURE CITED

This research aligns the new macro-finance model with the mainstream finance literature, using a latent variable with stochastic first and second moments to model the unit root. Because volatility depends upon the stochastic trend in this model, the Jensen effects induced by the convexity of the bond price function affect the associated slope parameter and not the intercept. This means that the trend affects the steady state inflation and spot rates without disturbing the asymptotic forward rate. The model was initially designed to tackle the asymptotic problem posed by the unit root, but in practice it seems that its superior performance stems from its ability to handle the heteroscedasticity of the macro-economic data rather than the asymptotic yield problem. Unfortunately, it is not possible to test these models on longer term yields using this historical data set because there have been long periods when the U.S. Treasury did not fund in the 20- to 30-year area. However, 30 year issuance has now resumed and the growing demand from pension providers is likely to keep this funding window open. Moreover, the increasing number of 50-year issues in the U.K. and French Treasury markets should generate data better suited to an empirical test of asymptotic model behavior.

In the meantime, the significance of the inflation-driven conditional heteroskedasticity found in U.S. macro-data motivates the use of the general affine model to study both the macroeconomy and the bond market. In contrast to the volatility-clustering effects implied by GARCH macro-models, this conditional heteroskedasticity is persistent, exhibiting a unit root. Mathematically, it is more tractable than the GARCH model, generating linear structures that could lend themselves not just to research on the term structure but to optimal control and similar intertemporal optimization problems. Empirically, this finding helps to explain the “Great Moderation”—the fall in output, interest rate, and inflation volatility seen since the mid 1980s (Bernanke 2004, Kim, Nelson, and Piger 2004)—attributing it to the fall in the inflation trend associated with the recession of the early 1980s. It reminds us that this so-called moderation is actually a return to the low-inflation, low-volatility epoch that characterized the early postwar years. The “general affine” macro-model helps to explain both the rise and subsequent fall in volatility.

Compared to the mainstream finance model of the bond market, the macro-finance model can use a relatively large number of factors (11) because the parameters of the model are informed by macroeconomic as well as yield data (with a total of 1,700 data points). It can also use an unrestricted specification of the price of risk, with a large number of parameters. It is particularly informative about the stochastic volatility parameters, identifying these with the volatility parameters of the macro-model, which are well determined. The relative adjustment speeds mean that the behavior of the yield curve is largely dictated by three factors: the inflation end-point, the real interest rate factor, and the spot rate. The model is consistent with the traditional three-factor finance specification in this sense but links these factors to the behavior of the macroeconomy. This research opens the way to new CBM-based studies of monetary policy and a much richer term structure specification, incorporating the best features of both macro-finance and mainstream finance models.

### Appendices

- Top of page
- Abstract
- 1. THE GENERAL AFFINE MACRO-MODEL FRAMEWORK
- 2. THE GENERAL AFFINE YIELD CURVE FRAMEWORK
- 3. THE EMPIRICAL MODELS
- 4. CONCLUSION
- Appendices
- LITERATURE CITED

#### APPENDIX A: THE STATE-SPACE REPRESENTATION OF THE MODEL

Define *x*_{t}={*y*′_{t}, *z*′_{t}}′; *v*_{t}={ɛ′_{t}, η′_{t}}′ and combine (1) and (2) to get:

- ((A1))

where

Stacking (A1) puts the system into state space form (4), where *X*_{t}={*y*′_{t}, *z*′_{t}, … , *z*′_{t-1}}′, *W*_{t}=*C*.{ɛ′_{t}, η′_{t}, 0_{1,N−k−n}}′ and:

- ((A2))

The second matrix repartitions Φ conformably with (10), so that Φ_{21} is (*N*− 1) × 1 and Φ_{22} is (*N*− 1)^{2}. Similarly:

where *C*_{21} is (*N*− 1) × 1 and *C*_{22} is (*N*− 1)^{2}. Comparing this with the partition of (A2), note that:

- ((A3))

Similarly for *x*_{t}={*x*_{1,t}, *x*′_{2,t}}′ repartition *v*_{t}={*w*_{1,t}, *v*′_{2,t}}′, *w*_{t}={*w*_{1,t}, *w*′_{2,t}}′ and (A1) conformably and write Γ as:

where Γ_{22} is a (*k*+*n*− 1)^{2} lower triangular matrix with unit diagonals and Γ_{21} is a (*k*+*n*− 1) column vector. The errors in *x*_{2,t+1} are decomposed into orthogonal components that are related to *w*_{1,t+1} and *v*_{2,t+1}=*s _{t}u*

_{2,t+1}:

- ((A4))

where *u*_{2,t+1} is an (*n*+*k*− 1) vector of standard normal variables, ; δ_{mj}≥ 0, *m*= 0, 1; *j*= 1, … , *k*+*n*, and *E*[*u*_{2,t+1}*w*_{1,t+1}]= 0_{(k+n−1)}; *u*_{2,t+1}∼*N*[0_{(k+n−1)}, *I*_{(k+n−1)}]. The error structure of (10) follows from (A4) as:

- ((A5))

- ((A6))

where *U*_{2,t+1}={(*u*′_{2,t+ 1}, 0′_{N-k-n}}′; ; *D*=*Diag*{1′_{k+n−1}, 0′_{N−k−n}}, so that *S _{t}D*=

*S*

_{t}. This implies the Gaussian conditional MGFs:

- ((A7))

where

- ((A8))

Finally, the conditional value *X*_{2,t+1} | *x*_{1,t+1} can be represented using (10), (A3), and (A5) as:

- ((A9))

#### APPENDIX B: AFFINE YIELD STRUCTURES

This appendix derives the MGF of the distribution under the risk neutral measure and shows how it can be employed as a “model generating function” to derive the yield model, forward rates, and risk premia as well as the moments of the macro-system under .

##### B.1 The Risk-Neutral Probability Measure

Measure adjusts the state probabilities using a multiplicative state-dependent subjective-utility weight *N*_{t+1} (with the logarithm *n*_{t+1}) so that the time *t* conditional risk neutral expectation of a scalar random variable *Z*_{t+1}:

- ((B1))

Bond (and other asset) prices are discounted expectations of future payoffs and prices defined under this measure:

- ((B2))

(Campbell, Lo, and MacKinlay 1997, Cochrane 2000). Recall that the MGF for measure is:

- ((B3))

Using (11) to replace *P*_{τ−1,t+1} in (B2) gives a similar form with ν=−Ψ_{τ−1}:

- ((B4))

where *J*_{r} is a selection vector such that *J*′_{r}*X*_{t}=*r*_{1,t}.

The discount factor is naturally exponential and if the MGF under the risk-neutral measure is exponential-affine in *X*_{t}, then so is the expectation. Thus, (B4) is of the form (11), with coefficients that are obtained recursively by matching the coefficients (for maturity τ− 1) of the state variables *X*_{t} in the exponents of (B4) with those (for τ) in (11). For this to be the case, *N*_{t+1} must be an exponential-affine function of the state variables or error terms:

- ((B5))

where λ_{1,t} is a scalar, and Λ_{2,t}=[λ′_{2,t}, 0′_{N−(k+n)}]′ is a (*N*− 1) × 1 deficient vector containing λ_{2,t}, which is a (*k*+*n*− 1) × 1 vector of coefficients related to the prices of risk associated with shocks to *x*_{2,t+1}. In the basic affine model class these coefficients are constant and variations in the risk premia only depend upon those in volatility. However, in the “essentially affine” specification of Duffee (2002) they are linear in *x*_{t}, allowing an addition source of variation in the premia. I adopt his specification using (13) with:

- ((B6))

where Λ′_{12}, Λ_{20} and Λ_{21} are (*N*− 1) × 1 vectors and Λ_{22} is an (*N*− 1)^{2} matrix of parameters to be estimated. The parameter λ_{11} allows *x*_{1} to influence the asymptotic risk premium through the price of risk. However, since *x*_{1} affects this through volatility it is redundant in and is set to zero. For the *EA*_{1} specifications M1 and M2 to be admissible it is also necessary that Λ_{12}= 0_{N−1} and to facilitate the encompassing tests I follow Dewachter and Lyrio (2006) and use this restriction for M0.

##### B.2 The MGF under the Risk-Neutral Measure

Using (B1) and (B3), the MGF of the distribution under the risk-neutral measure can be represented as: =*E*[exp [*n*_{t+1}+ν′*X*_{t+1}] | *X*_{t}]. Substituting (B5) and (A9) and noting that *x*_{1,t+1} and *U*_{2,t+1} are independent allows this to be factorized as:

- ((B7))

For the baseline model M2, these expectations are evaluated using (7) and (A7), substituting for :

This probability density is normalized to unity using: . Substituting this back and using (15), (B6), and setting λ_{1,t}=λ_{10} gives:

- ((B8))

Using this as a moment-generating function (differentiating w.r.t. {ν_{1}, ν_{2}} and setting these parameters to zero) gives (14). The formulae (16), (18), and (20) follow by substituting ν=−Ψ_{τ−1}, into (B8), substituting this into (B4) and equating the coefficients of *X*_{t} in the exponent with those in (11). M1 follows immediately from the restrictions noted in the main text.

##### B.3 The Standard EA_{0} Model as a Special Case

The standard way to obtain the moments and yield structure for the *EA*_{0} specification is to use (5) instead of (7) to evaluate the second expectation in (B7). However it is more instructive to derive these from the formulae for M2, taking the limit as *c* tends to infinity and setting Σ_{1} and δ_{11} to zero. This specializes the baseline parameters as shown in the final column of Table 2. We expand the denominator in as the geometric series [1 −λ_{10}/*c*+ (λ_{10}/*c*)^{2}−⋯] and use δ_{01}=μ/*c* to get the second-order approximation: , which is approximated arbitrarily closely by for large values of *c* (*o*(*c*^{−1}) denotes terms of order *c*^{−1} or smaller). follows from in the same way. Similarly, may be written as the second-order approximation: , which reduces to the first-order expansion upon the substitution of zero for the limiting value of as *c* tends to infinity. follows from in the same way.

Now consider the *EA*_{0} yield curve formulae. First, note that the recursive nature of the coefficient system means that , and *c* do not affect (16). Next, expand the denominators in (18) as above to give the quadratic approximation:

- ((B9))

Substituting the limit δ_{11}= 0 reduces this to a linear difference equation and then substituting (15), (A3), Σ_{1}= 0 and the coefficients of Table 2 gives:

- ((B10))

which reproduces (21) upon substitution of . Finally, to specialize the intercept (22) for *EA*_{0}, take a Taylor approximation of the logarithmic term in (20) around ln [1] use δ_{01}=μ/*c* and then expand the denominator terms.

Neglecting the terms *o*(*c*^{−1}) gives a second-order approximation. Substituting back into (20) using (15) and Table 2 gives (22):

- ((B11))

##### B.4 Risk Premia

The risk premia depend entirely upon the difference between the two measures and . To see this, note that the premium on a τ-period bond is the expected return, less the spot rate. The gross expected rate of return is the expected payoff *E*[*P*_{τ−1,t+1} |*X*_{t}] divided by its current price *P*_{τ,t}=. Taking the natural logarithm expresses this as a percentage return and subtracting the spot rate *r*_{1,t} then gives the risk premium: This is affine provided that the MGF is exponential-affine under both measures. Substituting (11), (7), and (B8) with ν=Ψ′_{τ−1} gives the risk premium. In the case of the *EA*_{1} unit root model this gives:

- ((B12))

The linear term on the first line is the compensation for the bond's exposure to shifts in *X*_{2,t}, which is negligible for a portfolio or security like an ultra-long bond, with a yield that mimics the asymptotic portfolio. The nonlinear term on the second line shows the premium on the asymptotic portfolio and is zero if λ_{10}= 0. It is positively related to *x*_{1,t} if λ_{10} < 0, and negatively related if λ_{10} > 0 since:

given (19). The risk premium in the *EA*_{0}(*N*) model is:

- ((B13))

Recall that this model requires the restriction λ_{11} > 0. This means that the associated factor risk premium (the asymptotic premium) is negatively related to the inflation trend. This premium is shown by δ_{01}(λ_{10}+λ_{11}*x*_{1,t}) in the second term. In practice, this effect has the effect of offsetting the effect of the component Ψ′_{2,τ−1}ϒ_{0}*x*_{1,t} shown in the first term (which is positive in the empirical model). Thus, the influence of the trend on the premia shown in Figure 9 is more pronounced in model M1.

##### B.5 Forward Rates and Asymptotic Behavior

Taking logs of (11) and maturity-differencing gives the affine forward rate structure:

- ((B14))

This shows that the asymptotic behavior of the forward rate depends critically upon whether the slope coefficients converge to constants. If so, the last term vanishes and Δγ_{τ+1} and hence the forward rate asymptote (*f**_{t}) is constant. Since Ψ*_{2}=lim_{τ∞}Ψ_{2,τ}= is constant, this just depends upon the behavior ψ_{1,τ}.Table 3 shows the asymptotic effect on the yield model of imposing unit root restrictions on *x*_{1,t} under . In *EA*_{1}, ψ_{1,τ} asymptotes to a constant, so β*_{1} and *f** are zero. However, (21) shows that in *EA*_{0}: lim_{τ∞}(ψ_{1,τ+1}−ψ_{1,τ})== Consequently, . This expression is equal to the asymptotic effect of *x*_{1,t} on *r*_{1,t} under and as such it should be close to unity. Substituting these coefficients into (22) and (B14) gives the asymptotic behavior of the forward rate reported in Table 3. The final term in this expression is the Jensen effect associated with ψ_{1,τ}*x*_{1,t}, which behaves like in the limit. These results illustrate the basic theorem of Dybvig, Ingersoll, and Ross (1996), which says that in an arbitrage-free framework “the limiting forward interest rate, if it exists, can never fall.” In other words the forward rate must either fail to converge with maturity (as in the unit root model), or must asymptote to a constant .

#### APPENDIX C: THE KALMAN FILTER AND THE LIKELIHOOD FUNCTION

These models were estimated and tested using the fminsearch algorithm on *Matlab*.^{14} In this model, the unobservable variables are modeled using the extended Kalman filter (Harvey 1989, Duffee and Stanton 2004). This method assumes that the revisions in (2) are approximately normally distributed:

where

I represent expectations conditional upon the available information with a “hat” (so that *y*_{s}; *s*≥*t*) and define the covariance matrices:

- ((C1))

where using (2):

- ((C2))

Similarly, using (1): , where

- ((C3))

and using (12): , where

- ((C4))

The *t*-conditional covariance matrix for this (*t*+ 1)-dated system is:

where ; *M*_{t}=*G*[*S*_{0}+*S*_{1}*y*_{t,t}]*G*′; *S*_{i}=*Diag*{δ^{2}_{i1} … , δ^{2}_{in}}; *i*= 0, 1 and is defined in (12). This allows the expectations to be updated as:

- ((C5))

- ((C6))

The (log) likelihood for period *t*+ 1 is thus:

- ((C7))

The log-likelihood for the full sample follows by iterating (C1), (C2), (C5), and (C6) forward given suitable starting values; substituting (C3) and (C4) then summing (C7) over *t*= 1, … ,*T*.

- 1
Ball (1992) offers a theoretical analysis of this phenomenon and the empirical evidence is examined by Brunner and Hess (1993), Holland (1995), Caporale and McKiernan (1997), and others. There is also an extensive literature on the effect of inflation and macro-economic volatility on the equity risk premium (Brandt and Wang 2003, Lettau, Ludvigson, and Wachter Forthcoming).

- 2
In this paper,

*Diag*{δ} represents a matrix with the elements of the row vector δ in the main diagonal and zeros elsewhere. 0_{a}is the (*a*× 1) × 1 zero vector, 1_{a}is the (*a*× 1) × 1 summation vector, 0_{a,b}is the (*a*×*b*) zero matrix, and*I*_{a}is the*a*^{2}identity matrix. - 3
I am grateful to Tony Rodrigues of the New York Federal Reserve Bank for supplying a copy of this yield data set.

- 4
Preliminary tests showed no significant evidence of autoregressive conditional heteroskedasticity (ARCH) in this data set.

- 5
This parameter is equivalent to ϕ

_{1}in (3) in this case. - 6
In the variable driving volatility

*x*_{1,t}has a noncentral χ^{2}distribution and is nonnegative, keeping the variance structure Σ_{0}+Σ_{1}*x*_{1,t−1}for*X*_{2,t}nonnegative definite. - 7
The usual convention in macro-finance models is that this represents measurement error.

- 8
A shift in the parameters from (4) to (14) could occur either because of the risk adjustment implied by (B5) or because the bond market does not use the true values of the parameters in (4). However, it is not possible to distinguish these two effects without imposing restrictions on (B5).

- 9
Their model uses a Gaussian approximation to the CIR process describing the spot rate, due originally to Sun (1992).

- 10
Substituting (17) into (18) gives ψ*

_{1}as the solution to: ψ*_{1}=−−Ψ*′_{2}ϒ_{1}− This may be arranged as: 0 =ϑ^{2}+ϑ (*c*(1 −ξ_{1}) −ζ) −*c*ζ, where ϑ=ψ*_{1}+λ_{10}+Ψ*′_{2}*C*_{21}, and The intercept term*c*ζ shows the product of the roots and is a very large negative number. Consequently, one root is a large negative and the other a large positive number. Phase analysis reveals that the recursion (18) selects the positive root. - 11
These are , δ

_{01}, δ_{11},**Δ**_{0}(4),**Δ**_{1}(4),*H*_{1}(4), ϒ_{1}(3), Λ_{22}(13),*G*(3), Φ(27), and ϕ(2). Estimates are reported in Table 4. It was found that although was significant (Table 4c) the remaining elements of the first rows of Λ_{22}and ϒ_{1}(or ϒ_{0}) were very poorly determined and could be eliminated without significantly reducing the likelihood. The structural parameters μ and*c*follow from (8) given ξ_{1},δ_{01}, and δ_{11}. - 12
This restriction is imposed via (8) by setting δ

_{01}= 0. - 13
However, the imaginary components of the first root is small, meaning that the macro-model is dominated by exponential adjustment effects.

- 14
I am very grateful to Zhuoshi Liu for suggesting this algorithm and converting my basic files from

*Mathematica*to*Matlab*.

### LITERATURE CITED

- Top of page
- Abstract
- 1. THE GENERAL AFFINE MACRO-MODEL FRAMEWORK
- 2. THE GENERAL AFFINE YIELD CURVE FRAMEWORK
- 3. THE EMPIRICAL MODELS
- 4. CONCLUSION
- Appendices
- LITERATURE CITED

- 1996) “Testing Continuous Time Models of the Spot Interest Rate. Review of Financial Studies, 9, 385–42. . (
- 2003) “A No-Arbitrage Vector Autoregression of Term Structure Dynamics with Macroeconomic and Latent Variables. Journal of Monetary Economics, 43, 75–87. , and . (
- 1992) “Why Does Inflation Raise Inflation Uncertainty Journal of Monetary Economics, 29, 371–88. . (
- 2004) “The Great Moderation.” In
*Proceedings of the Meetings of the Eastern Economic Association*. Washington , DC , February. http://www.federalreserve.gov/BOARDDOCS/SPEECHES/2004/20040220/default.htm . ( - 2003) “Time Varying Risk Aversion and Unexpected Inflation. Journal of Monetary Economics, 50, 1457–98. , and . (
- 1993) “Are Higher Levels of Inflation Less Predictable? A State Dependent Conditional Heteroscedasticity Approach. Journal of Business & Economic Statistics, 11, 187–97. , and . (
- 1997) “
*The Econometrics of Financial Markets*.” Princeton , NJ : Princeton University Press. , and and . ( - 1997) “High and Variable Inflation: Further Evidence on the Friedman Hypothesis. Economics Letters, 54, 65–8. , and . (
- 1992) “An Empirical Comparison of Alternative Models of the Short Term Interest Rate. The Journal of Finance, 47, 1209–27. , , , and . (
- 1993) “Maximum Liklihood Estimation for a Multifactor Equilibrium Model of the Term Structure of Interest Rates. Journal of Fixed Income, 3, 14–31. , and . (
- 1997) “Time Series for Economics and Finance.” Unpublished lecture notes. http://www-gsb.uchicago.edu/fac/john.cochrane/research/Papers/timeser1.pdf . (
- 2000) “Time Series for Economics and Finance.” http://www.gsb.uchicago.edu/fac/john.cochrane/research/Papers/timeser1.pdf (
- 1985) “A Theory of the Term Structure of Interest Rates. Econometrica, 52, 385–407. , , and . (
- 2000) “Specification Analysis of Affine Term Structure Models. Journal of Finance, 55, 415–41. , and . (
- 2002) “Expectation Puzzles, Time-Varying Risk Premia and Affine Models of the Term Structure. Journal of Financial Economics, 63, 415–41. , and . (
- 2006) “Macro Factors and the Term Structure of Interest Rates. Journal of Money, Credit and Banking, 38, 119–40. , and . (
- 2006) “A Joint Model for the Term Structure of Interest Rates and the Macroeconomy. Journal of Applied Econometrics, 21, 439–26. , , and . (
- 2002) “Term Premia and Interest Rate Forecasts in Affine Models. Journal of Finance, 62, 405–43. (
- 2004) “Estimation of Dynamic Term Structure Models.” Mimeo, Haas School, UC Berkeley. , and . (
- 2003) “Affine Processes and Applications in Finance. Annals of Applied Probability, 13, 984–1053. , , and . (
- 1996) “A Yield-Factor Model of Interest Rates.
*Mathematical Finance*, 6, 379–406. , and . ( - 1996) “Long Forward and Zero-Coupon Rates Can Never Fall. Journal of Business, 69, 121–43. , , and . (
- 1982) “Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of UK Inflation. Econometrica, 50, 987–1010. (
- 2006) “The Behaviour of Interest Rates. Review of Financial Studies, 45, 359–79 (
- 1977) “Nobel Lecture: Inflation and Unemployment. Journal of Political Economy, 85, 451–72. . (
- 2002) “Autoregressive Gamma Processes.” Working Paper, CREST. , and . (
- 1989)
*Forecasting, Structural Time Series Models and the Kalman Filter*. Cambridge : Cambridge University Press. . ( - 1995) Inflation and Uncertainty: Tests for Temporal Ordering. Journal of Money, Credit and Banking, 27, 827–37. . (
- 1970)
*Distributions in Statistics: Continuous Univariate Distributions*. Boston , MA : Houghton Miffin. , and . ( - 2004) “The Less Volatile U.S. Economy: A Bayesian Investigation of Timing, Breadth, and Potential Explanations. Journal of Business and Economic Statistics, 22, 80–93. , , and . (
- 2001) “Shifting Endpoints in the Term Structure of Interest Rates. Journal of Monetary Economics, 47, 613–52. , and . (
- 2005) “Permanent and Transitory Policy Shocks in an Empirical Macro Model with Asymmetric Information. Journal of Economic Dynamics and Control, 11, 1985–2015. , and . (
- 1992) “Testing the Null Hypothesis of Stationarity Against the Alternative of a Unit Root. Journal of Econometrics, 54, 159–78. , , , and . (
- The Declining Equity Premium: What Role Does Macroeconomic Risk Play Review of Financial Studies. , , and . (Forthcoming) “
- 1991) “US Term Structure Data.” Manuscript, Ohio State University . , and . (
- 1971) “The Mirage of Steady Inflation.
*Brookings Papers on Economic Activity*, 2, 485–98. . ( - 2002) “Assessing Nominal Income Rules for Monetary Policy with Model and Data Uncertainty. The Economic Journal, 112, 402–32. . (
- 2003) “A Macro-Finance Model of the Term Structure, Monetary Policy and the Economy.” FRBSF Working Paper, 2003–17. , and . (
- 1979) “The Non-Central Chi-Squared Distribution with Zero Degrees of Freedom. Biometrica, 381–86. . (
- 1992) “Interpreting the Macroeconomic Time Series Facts: The Effect of Monetary Policy. European Economic Review, 36, 975–1011. . (
- 1999) “Output Gap Uncertainty: Does It Matter for the Taylor Rule?” In
*Monetary Policy under Uncertainty*, edited by HuntBarry and OrrAndrew, pp. 10–29. Wellington : Reserve Bank of New Zealand . . ( - 1997) “A Non-Parametric Model of Term Structure Dynamics and the Market Price of Interest Rate Risk. Journal of Finance, 52, 1973–2002. . (
- 1992) “Real and Nominal Interest Rates: A Discrete-Time Model and Its Continuous Time Limit. Review of Financial Studies, 5, 581–611. . (
- 1999) “Inflation Targetting as a Monetary Policy Rule. Journal of Monetary Economics, 43, 607–54. . (