We propose a new method of testing stochastic dominance that improves on existing tests based on the standard bootstrap or subsampling. The method admits prospects involving infinite as well as finite dimensional unknown parameters, so that the variables are allowed to be residuals from nonparametric and semiparametric models. The proposed bootstrap tests have asymptotic sizes that are less than or equal to the nominal level uniformly over probabilities in the null hypothesis under regularity conditions. This paper also characterizes the set of probabilities that the asymptotic size is exactly equal to the nominal level uniformly. As our simulation results show, these characteristics of our tests lead to an improved power property in general. The improvement stems from the design of the bootstrap test whose limiting behavior mimics the discontinuity of the original test’s limiting distribution.
In this note we propose a simple method of measuring directional predictability and testing for the hypothesis that a given time series has no directional predictability. The test is based on the correlogram of quantile hits. We provide the distribution theory needed to conduct inference, propose some model free upper bound critical values, and apply our methods to stock index return data. The empirical results suggests some directional predictability in returns especially in mid range quantiles like 5%-10%.
We propose a procedure for estimating the critical values of the Klecan, McFadden, and McFadden (1990) test for first and second order stochastic dominance in the general k-prospect case. Our method is based on subsampling bootstrap. We show that the resulting test is consistent. We allow for correlation amongst the prospects and for the observations to be autocorrelated over time. Importantly, the prospects may be the residuals from certain conditional models.
We introduce a new method for the estimation of discount functions, yield curves and forward curves for coupon bonds. Our approach is nonparametric and does not assume a particular functional form for the discount function although we do show how to impose various important restrictions in the estimation. Our method is based on kernel smoothing and is defined as the minimum of some localized population moment condition. The solution to the sample problem is not explicit and our estimation procedure is iterative, rather like the backfitting method of estimating additive nonparametric models. We establish the asymptotic normality of our methods using the asymptotic representation of our estimator as an infinite series with declining coefficients. The rate of convergence is standaxd for one dimensional nonparametric regression.
We derive the asymptotic distribution of a new backfitting procedure for estimating the closest additive approximation to a nonparametric regression function. The procedure employs a recent projection interpretation of popular kernel estimators provided by Mammen et al. (1997), and the asymptotic theory of our estimators is derived using the theory of additive projections reviewed in Bickel et al. (1995). Our procedure achieves the same bias and variance as the oracle estimator based on knowing the other components, and in this sense improves on the method analyzed in Opsomer and Ruppert (1997). We provide ‘high level’ conditions independent of the sampling scheme. We then verify that these conditions are satisfied in a time series autoregression under weak conditions.
We develop stochastic expansions with remainder oP(n–2µ), where 0 < µ < 1/2, for a standardised semiparametric GLS estimator, a standard error, and a studentized statistic, in the linear regression model with heteroskedasticity of unknown form. We calculate the second moments of the truncated expansion, and use these approximations to compare two competing estimators and to define a method of bandwidth choice.
Keywords: Semiparametric estimation, GLS, Heteroskedasticity, Local linear regression, Asymptotic expansions
We propose a nonparametric empirical distribution function based test of an hypothesis of conditional independence between variables of interest. This hypothesis is of interest both for model specification purposes, parametric and semiparametric, and for non-model based testing of economic hypotheses. We allow for both discrete variables and estimated parameters. The asymptotic null distribution of the test statistic is a functional of a Gaussian process. A bootstrap procedure is proposed for calculating the critical values. Our test has power against alternatives at distance n-1/2 from the null; this result holding independently of dimension. Monte Carlo simulations provide evidence on size and power. Finally, we invert the test statistic to provide a method for estimating the parameters identified through the conditional independence restriction. They are asymptotically normal at rate root-n.
We develop order T-1 asymptotic expansions for the quasi-maximum likelihood estimator (QMLE) and a two step approximate QMLE in the GARCH(1,1) model. We calculate the approximate mean and skewness and hence the Edgeworth-B distribution function. We suggest several methods of bias reduction based on these approximation.
Existing specification tests for conditional heteroskedasticity are derived under the assumption that the density of the innovation, or standardized error, is Gaussian, despite the fact that many recent empirical studies provide evidence that this density is not Gaussian. We obtain specification tests for conditional heteroskedasticity under the assumption that the innovation density is a member of a general family of densities. Our test statistics maximize asymptotic local power and weighted average power criteria for the general family of densities. We establish both first order and second order theory for our procedures. Monte Carlo simulations indicate that asymptotic power gains are achievable in finite samples. We apply the tests to shock futures data sampled at high frequency and find evidence of conditional heteroskedasticity in the residuals from a GARCH(1,1) model, indicating that the standard (1,1) specification is not adequate.
We examine the higher order asymptotic properties of semiparametric regression estimators that were obtained by the general MINPIN method described in Andrews (1989). We derive an order n–1 stochastic expansion and give a theorem justifying order n – 1 distributional approximation of the Edgeworth type.
We examine the second order properties of various quantities of interest in the partially linear regression model. We obtain a stochastic expansion with remainder oP(n -2µ), where µ < 1/2, for the standardized semiparametric least squares estimator, a standard error estimator, and a studentized statistic. We use the second order expansions to correct the standard error estimates for second order effects, and to define a method of bandwidth choice. A Monte Carlo experiment provides favorable evidence on our method of bandwidth choice.
Keywords: Semiparametric estimation, Partially linear regression, Kernel, Local polynomial, Second order approximations, Bandwidth choice, Asymptotic expansions
We construct efficient estimators of the identifiable parameters in a regression model when the errors follow a stationary parametric ARCH(P) process. We do not assume a functional form for the conditional density of the errors, but do require that it be symmetric about zero. The estimators of the mean parameters are adaptive in the sense of Bickel [2]. The ARCH parameters are not jointly identifiable with the error density. We consider a reparameterization of the variance process and show that the identifiable parameters of this process are adaptively estimable.