Results 1  10
of
20
Resurrecting the (C)CAPM: A CrossSectional Test When Risk Premia Are TimeVarying
 Journal of Political Economy
, 2001
"... This paper explores the ability of conditional versions of the CAPM and the consumption CAPM—jointly the (C)CAPM—to explain the cross section of average stock returns. Central to our approach is the use of the log consumption–wealth ratio as a conditioning variable. We demonstrate that such conditio ..."
Abstract

Cited by 139 (5 self)
 Add to MetaCart
This paper explores the ability of conditional versions of the CAPM and the consumption CAPM—jointly the (C)CAPM—to explain the cross section of average stock returns. Central to our approach is the use of the log consumption–wealth ratio as a conditioning variable. We demonstrate that such conditional models perform far better than unconditional specifications and about as well as the FamaFrench threefactor model on portfolios sorted by size and booktomarket characteristics. The conditional consumption CAPM can account for the difference in returns between lowbooktomarket and highbooktomarket portfolios and exhibits little evidence of residual size or booktomarket effects. We are grateful to Eugene Fama and Kenneth French for graciously providing the
Expectations Hypotheses Tests
 Journal of Finance
, 2001
"... We investigate the Expectations Hypotheses of the term structure of interest rates and of the foreign exchange market using vector autoregressive methods for the U.S. dollar, Deutsche mark, andBritishpoundinterestratesandexchangerates. InadditiontostandardWaldtests,we formulate Lagrange Multiplier a ..."
Abstract

Cited by 17 (1 self)
 Add to MetaCart
We investigate the Expectations Hypotheses of the term structure of interest rates and of the foreign exchange market using vector autoregressive methods for the U.S. dollar, Deutsche mark, andBritishpoundinterestratesandexchangerates. InadditiontostandardWaldtests,we formulate Lagrange Multiplier and Distance Metric tests which require estimation under the nonlinear constraints of the null hypotheses. Estimation under the null is achieved by iterating on approximate solutions that require only matrix inversions. We use a biascorrected, constrained vector autoregression as a data generating process and construct extensive Monte Carlo simulations of the various test statistics under the null hypotheses. Wald tests suffer from severe size distortions and use of the asymptotic critical values results in gross overrejection of the null. The Lagrange Multiplier tests slightly underreject the null, and the Distance Metric tests overreject. Use of the small sample distributions of the different tests leads to a common interpretation of the validity of the Expectations Hypotheses. The evidence against the Expectations Hypotheses for these interest rates and exchange rates is much less strong than under asymptotic inference. 2 According to the Expectations Hypothesis, information in current interest rates provides the
The Term Structure with Highly Persistent Interest Rates
, 1998
"... If yields are assumed to have an exact unitroot, it has previously been shown that the rational expectations hypothesis of the term structure (REHTS) has been rejected by singleequation tests. However, small deviations from exact unitroot produce substantial changes in the small sample distributi ..."
Abstract

Cited by 7 (1 self)
 Add to MetaCart
If yields are assumed to have an exact unitroot, it has previously been shown that the rational expectations hypothesis of the term structure (REHTS) has been rejected by singleequation tests. However, small deviations from exact unitroot produce substantial changes in the small sample distributions of those tests and the normal approximation is no longer satisfactory. We assume that the yield of 1period zerocoupon bond follows a localtounity process with parameter c (c=0 for exact unit root) and use asymptotics to derive alternative distributions, which are far better approximations to ...finite sample distributions. Those asymptotic distributions depend crucially on c, and that allows us to analyze the impact of small deviations from unitroot on the distribution of the tests. Interestingly, for small values of c, the results obtained in the data do not imply a rejection of the REHTS. The above results are useful only when c is known or consistently estimable. Thus, the REHTS ...
Estimation of dsge models when the data are persistent
, 2007
"... Dynamic Stochastic General Equilibrium (DSGE) models are often solved and estimated under specific assumptions as to whether the exogenous variables are difference or trend stationary. However, even mild departures of the data generating process from these assumptions can severely bias the estimates ..."
Abstract

Cited by 7 (0 self)
 Add to MetaCart
Dynamic Stochastic General Equilibrium (DSGE) models are often solved and estimated under specific assumptions as to whether the exogenous variables are difference or trend stationary. However, even mild departures of the data generating process from these assumptions can severely bias the estimates of the model parameters. This paper proposes new estimators that do not require researchers to take a stand on whether shocks have permanent or transitory effects. These procedures have two key features. First, the same filter is applied to both the data and the model variables. Second, the filtered variables are stationary when evaluated at the true parameter vector. The estimators are approximately normally distributed not only when the shocks are mildly persistent, but also when they have near or exact unit roots. Simulations show that these robust estimators perform well especially when the shocks are highly persistent yet stationary. In such cases, linear detrending and first differencing are shown to yield biased or imprecise estimates.
Finite Sample Properties of EMM, GMM, QMLE, and MLE for a SquareRoot Interest Rate Diffusion Model
, 2001
"... This paper performs a Monte Carlo study on Ecient Method of Moments (EMM), Generalized Method of Moments (GMM), QuasiMaximum Likelihood Estimation (QMLE), and Maximum Likelihood Estimation (MLE) for a continuoustime squareroot model under two challenging scenarios  high persistence in mean and s ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
This paper performs a Monte Carlo study on Ecient Method of Moments (EMM), Generalized Method of Moments (GMM), QuasiMaximum Likelihood Estimation (QMLE), and Maximum Likelihood Estimation (MLE) for a continuoustime squareroot model under two challenging scenarios  high persistence in mean and strong conditional volatility  that are commonly found in estimating the interest rate process. MLE turns out to be the most eficient of the four methods, but its finite sample inference and convergence rate suffer severely from approximating the likelihood function, especially in the scenario of highly persistent mean. QMLE comes second in terms of estimation efficiency, but it is the most reliable in generating inferences. GMM with lagaugmented moments has overall the lowest estimation efficiency, possibly due to the ad hoc choice of moment conditions. EMM shows an accelerated convergence rate in the high volatility scenario, while its overrejection bias in the mean persistence scenario is unacceptab...
Can We Really Observe Hyperbolic Discounting?
, 2002
"... This paper proposes a new, more robust, experiment to test for the presence of hyperbolic discounting. Recently, a growing literature has studied intertemporal choice when individuals discount the future hyperbolically. These preferences generate dynamically inconsistent choices, in contrast with th ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
This paper proposes a new, more robust, experiment to test for the presence of hyperbolic discounting. Recently, a growing literature has studied intertemporal choice when individuals discount the future hyperbolically. These preferences generate dynamically inconsistent choices, in contrast with the usual assumption of exponential discounting, where this issue cannot arise. Hyperbolic discounting is justified based on experimental evidence of individual selfcontrol problems. We argue that this interpretation depends crucially on the absence of uncertainty. We show that, once uncertainty is included, the observed behavior is compatible with exponential discounting. We then test for the presence of hyperbolic discounting in a new experiment that controls for uncertainty. The experiment offers two choice sets, the second being a strict subset of the first. Exponential discounters will (possibly weakly) prefer the largest one. Hyperbolic discounters, in contrast, will (strictly) prefer the second set because its design makes it equivalent to a commitment technology. The experiment is conducted...
Improvement in Finite Sample Properties of the HansenJagannathan Distance Test ∗
, 2007
"... Jagannathan and Wang (1996) derive the asymptotic distribution of the HansenJagannathan distance (HJdistance) proposed by Hansen and Jagannathan (1997), and develop a specification test of asset pricing models based on the HJdistance. While the HJdistance has several desirable properties, Ahn a ..."
Abstract
 Add to MetaCart
Jagannathan and Wang (1996) derive the asymptotic distribution of the HansenJagannathan distance (HJdistance) proposed by Hansen and Jagannathan (1997), and develop a specification test of asset pricing models based on the HJdistance. While the HJdistance has several desirable properties, Ahn and Gadarowski (2004) find that the specification test based on the HJdistance overrejects correct models too severely in commonly used sample size to provide a valid test. This paper proposes to improve the finite sample properties of the HJdistance test by applying the shrinkage method (Ledoit and Wolf, 2003) to compute its weighting matrix. The proposed method improves the finite sample performance of the HJdistance test significantly. JEL classification: C13; C52; G12
On Testing Overidentifying Restrictions in Dynamic Panel Data Models
, 2000
"... The conventional Sargan (1958) / Hansen (1982) test of overidentifying restrictions and the Tilting Parameter test of Imbens, Spady and Johnson (1998) are compared in the context of the AR(1) dynamic panel data model using Monte Carlo experiments. Interestingly, the size properties of the former ar ..."
Abstract
 Add to MetaCart
The conventional Sargan (1958) / Hansen (1982) test of overidentifying restrictions and the Tilting Parameter test of Imbens, Spady and Johnson (1998) are compared in the context of the AR(1) dynamic panel data model using Monte Carlo experiments. Interestingly, the size properties of the former are found to be superior in this setting. Nevertheless, the Sargan / Hansen test is found to have no power in panels of dimensions that are commonly encountered in empirical work. A simple procedure for reducing the number of moment conditions tested is shown to improve both the size and power properties of the conventional test. A useful diagnostic procedure is also suggested. JEL Classification: C23; C12. Keywords:
FINITESAMPLE INFERENCE FOR STATIONARY AND NONSTATIONARY AUTOREGRESSIVE PROCESSES
, 2000
"... Ce cahier a également été publié par le Centre interuniversitaire de recherche en économie quantitative (CIREQ) sous le numéro 122000. ..."
Abstract
 Add to MetaCart
Ce cahier a également été publié par le Centre interuniversitaire de recherche en économie quantitative (CIREQ) sous le numéro 122000.
Federal Reserve Bank of New York Forthcoming in Journal of Political Economy
, 2001
"... data on ĉay can be found at ..."