Yiguo Sun

Semiparametric estimation of fixed-effects panel data varying coefficient models

We consider the problem of estimating a varying coefficient panel data model with fixed-effects (FE) using a local linear regression approach. Unlike first-differenced estimator, our proposed estimator removes FE using kernel-based weights. This results a one-step estimator without using the backfitting technique. The computed estimator is shown to be asymptotically normally distributed. A modified least-squared cross-validatory method is used to select the optimal bandwidth automatically. Moreover, we propose a test statistic for testing the null hypothesis of a random-effects varying coefficient panel data model against an FE one. Monte Carlo simulations show that our proposed estimator and test statistic have satisfactory finite sample performance.

International correlations across stock markets and industries: trends and patterns 1988–2002

Data from eight major stock markets world-wide and five industries in each market are analysed. The correlations of return indices between countries and industries are studied with the hope of finding answers or confirming previous empirical answers to the following questions and the implications of these findings for investment strategy determined. (1) Do both the country-specific correlations and industry-specific correlations fluctuate significantly over time between 1988 and 2002? (2) Are the country-specific and industry-specific correlations positively related to stock market volatilities? It is concluded that: First, the correlations among national stock markets have been increasing between 1988 and 2002 and the correlations are not constant over the time period of this research. This indicates that the effect of globalization outweighs country-specific factors in determining the co-movements of the markets. Second, the correlations are positively related to volatility in the stock markets in this sample. Correlations rise in periods when conditional volatility of markets is large. Finally, in most cases, correlations between national stock markets are greater than those between the five industries chosen in these markets, indicating that investment diversification across industries provides greater risk reduction benefits than diversification across countries.

A Consistent Nonparametric Equality Test of Conditional Quantile Functions

This paper proposes a consistent nonparametric test for testing for equality of unknown conditional quantile curves across subgroups within a survey data framework. Moreover, to improve the small-sample performance of the test, we propose a modified version of wild bootstrap procedure in a quantile context. Monte Carlo evidence shows that the performance of the test in small samples is adequate but is improved significantly when the bootstrap critical values are used.I thank the co-editor and two referees for helpful comments that improved the paper. Financial support from the Social Sciences and Humanities Research Council of Canada is gratefully acknowledged.

SEMIPARAMETRIC FUNCTIONAL COEFFICIENT MODELS WITH INTEGRATED COVARIATES

Cai, Li, and Park ( Journal of Econometrics , 2009) and Xiao ( Journal of Econometrics , 2009) developed asymptotic theories for estimators of semiparametric varying coefficient models when regressors are integrated processes but the smooth coefficients are functionals of stationary processes. Using a recent result from Phillips ( Econometric Theory , 2009), we extend this line of research by allowing for both the regressors and the covariates entering the smooth functionals to be integrated variables. We derive the asymptotic distribution for the proposed semiparametric estimator. An empirical application is presented to examine the purchasing power parity hypothesis between U.S. and Canadian dollars.

Data-Driven Bandwidth Selection for Nonstationary Semiparametric Models

This article extends the asymptotic results of the traditional least squares cross-validatory (CV) bandwidth selection method to semiparametric regression models with nonstationary data. Two main findings are that (a) the CV-selected bandwidth is stochastic even asymptotically and (b) the selected bandwidth based on the local constant method converges to 0 at a different speed than that based on the local linear method. Both findings are in sharp contrast to existing results when working with weakly dependent or independent data. Monte Carlo simulations confirm our theoretical results and show that the automatic data-driven method works well.

A CONSISTENT NONPARAMETRIC TEST ON SEMIPARAMETRIC SMOOTH COEFFICIENT MODELS WITH INTEGRATED TIME SERIES

In this paper, we propose a simple nonparametric test for testing the null hypothesis of constant coefficients against nonparametric smooth coefficients in a semiparametric varying coefficient model with integrated time series. We establish the asymptotic distributions of the proposed test statistic under both null and alternative hypotheses. Moreover, we derive a central limit theorem for a degenerate second order U-statistic, which contains a mixture of stationary and nonstationary variables and is weighted locally on a stationary variable. This result is of independent interest and useful in other applications. Monte Carlo simulations are conducted to examine the finite sample performance of the proposed test.

Estimation and Inference in Functional-Coefficient Spatial Autoregressive Panel Data Models with Fixed Effects

This paper develops an innovative way of estimating a functional-coefficient spatial autoregressive panel data model with unobserved individual effects which can accommodate (multiple) time-invariant regressors in the model with a large number of cross-sectional units and a fixed number of time periods. The methodology we propose removes unobserved fixed effects from the model by transforming the latter into a semiparametric additive model, the estimation of which however does not require the use of backfitting or marginal integration techniques. We derive the consistency and asymptotic normality results for the proposed kernel and sieve estimators. We also construct a consistent nonparametric test to test for spatial endogeneity in the data. A small Monte Carlo study shows that our proposed estimators and the test statistic exhibit good finite-sample performance.

The Value-Undermining Effects of Rock Mining on Nearby Residential Property: A Semiparametric Spatial Quantile Autoregression

Rock mining operations, including limestone and gravel production, have considerable adverse effects on residential quality of life due to elevated noise and dust levels resulting from dynamite blasting and increased truck traffic. This paper provides the first estimates of the effects of rock mining — an environmental disamenity — on local residential property values. We focus on the relationship between a house’s price and its distance from nearby rock mine. Our analysis studies Delaware County, Ohio which, given its unique features, provides a natural environment for the valuation of property-value-suppressing effects of rock mines on nearby houses. We improve upon the conventional approach to valuating adverse effects of environmental disamenities based on hedonic house price functions. Specifically, in a pursuit of robust estimates, we develop a novel (semiparametric) partially linear spatial quantile autoregressive model which accommodates unspecified nonlinearities, distributional heterogeneity as well as spatial dependence in the data. We derive the consistency and normality limit results for our estimator as well as propose a consistent model specification test. We find statistically and economically significant propertyvalue-suppressing effects of being located near an operational rock mine which gradually decline to insignificant near-zero values at a roughly ten-mile distance. Our estimates suggest that, all else equal, a house located a mile closer to a rock mine is priced, on average, at about 2.3–5.1% discount, with more expensive properties being subject to larger markdowns.

Endogeneity in Semiparametric Threshold Regression

In this paper, we investigate semiparametric threshold regression models with endogenous threshold variables based on a nonparametric control function approach. Using a series approximation we propose a two-step estimation method for the threshold parameter. For the regression coefficients, we consider least-squares estimation in the case of exogenous regressors and two-stage least-squares estimation in the case of endogenous regressors. We show that our estimators are consistent and derive their asymptotic distribution for weakly dependent data. Furthermore, we propose a test for the endogeneity of the threshold variable, which is valid regardless of whether the threshold effect is zero or not. Finally, we assess the performance of our methods using a Monte Carlo simulation.

Nonparametric Models with Random Effects

This chapter considers the three-dimensional nonparametric models with random effects, and proposes pooled local linear and two-step estimators for them. We find that the pooled local linear estimator can be inconsistent when the sum of all the error term covariances in absolute values diverges to infinity too quickly. When the pooled local linear estimator is consistent, the optimal convergence rate of the estimator, its corresponding optimal bandwidth and asymptotic variance depend on the number of regressors and the limit of certain sample indices ratio; and we propose an asymptotically more efficient two-step estimator along the line of Su et al. (2013). Some extensions on nonparametric models with fixed effects, mixed effects, and higher dimensions are also discussed.