Bryan Campbell

Exact Nonparametric Tests of Orthogonality and Random Walk in the Presence of a Drift Parameter

In this paper, finite-sample nonparametric tests of conditional independence and random walk are extended to allow for an unknown drift parameter. The tests proposed are based on simultaneous inference methods and remain exact in the presence of general forms of feedback, nonnormality and heterskedasticity. Further, in two simulation studies, the authors confirm that the nonparametric procedures are reliable and find that they display power comparable or superior to that of conventional tests. Copyright 1997 by Economics Department of the University of Pennsylvania and the Osaka University Institute of Social and Economic Research Association. (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.) (This abstract was borrowed from another version of this item.)

Federal budget projections: a nonparametric assessment of bias and efficiency

As an important initial step in the annual budget process, the President presents to Congress each January his budget with details of federal spending activity and priorities. Our paper is a statistical assessment of the merit of the budget figures submitted to Congress. We investigate the overall budget as well as several important specific accounts. An important aspect of our paper is the introduction of a nonparametric methodology which incorporates exact tests for assessing the unbiasedness, and the internal and external consistency of forecasts. The empirical evidence shows that the nonparametric results confirm the presence of bias in forecasts on the outlay side suggested by regression results, but tends to find fewer series exhibiting bias on the revenue side. On the other hand the nonparametric approach lends greater support to the conclusion that the governments budget projections do not fully exploit available information. Copyright 1995 by MIT Press.

Over-rejections in rational expectations models: A non-parametric approach to the Mankiw-Shapiro problem☆

Abstract Standard tests for rationality may not have the correct size if there is feedback from innovations to future values of the regressors. It is shown that nonparametric tests reject at their nominal level and display good power in a variety of specifications of a model involving feedback.

Nonparametric Tests of the Unbiasedness of Olympic Figure-Skating Judgments

We use recently developed nonparametric tests for orthogonality to test the null hypothesis of an absence of national bias in the judging of Olympic figure-skating events. The results, and accompanying parametric estimates, offer strong evidence of a small bias which changes little across time. There is also some evidence that the bias is more marked for skaters who are medal contenders than for less strong competitors.

Inference in expectations models of the term structure: A non-parametric approach

Recent research has examined apparent deviations from the expectations theory of the term structure detectable in regression tests, which may be interpreted as efficiency tests. Efficiency is rejected in many studies. Inference is complicated, however, by the non-normality of regression residuals, invalidating standard parametric test procedures. The present paper examines these rejections using robust diagnostic methods and non-parametric tests. We find some evidence against the expectations theory of the term structure in U.S. data, but not in Canadian. We also investigate the possible explanation of a link between forecast error and the yield spread through models of time-variation in the liquidity premium.

Non-Parametric Regression Models of Deviations from Othogonality in the Expectations Theory of the Term Structure

Regression tests of the expectations theory of the term structure typically reject the null hypothesis of orthogonality between implied forecast errors and the yield spreads. In the statistical literature on the term structure, these rejections are sometimes attributed to time-varying liquidity premia, and Engle et al. (1987) suggest that the ARCH-M model of time-variation in the liquidity premium may be sufficient to account for rejections of the expectations theory. We use non-parametric (kernel) regression to explore the regression test results on a number of data sets, and find some evidence of a persistent deviation from orthogonality for large absolute values of the spread. Incorporating ARCH-in-mean into models of the term premium indicates that this specification does explain significant time variation in liquidity premia, but the effect does not apepar to be sufficient to account for all of the deviations from orthogonality of forecast errors and spreads.