This paper considers nonparametric estimation of smooth conditional distribution functions (CDFs) using kernel smoothing methods. We propose estimation using a new smoothed local linear (SLL) estimator. Estimation bias is reduced through the use of a local linear estimator rather than local averaging. Estimation variance is reduced through the use of smoothing. Asymptotic analysis of mean integrated squared error (MISE) reveals the form of these efficiency gains, and their magnitudes are demonstrated in numerical simulations. Considerable attention is devoted to the development of a plugin rule for bandwidths which minimize estimates of the asymptotic MISE. We illustrate the estimation method with an application to the U.S. quarterly GDP growth rate. ∗Research supported by the National Science Foundation. †Department of Economics, 1180 Observatory Drive, University of Wisconsin, Madison, WI 53706
[1]
M. Wand,et al.
EXACT MEAN INTEGRATED SQUARED ERROR
,
1992
.
[2]
Simon J. Sheather,et al.
Using non stochastic terms to advantage in kernel-based estimation of integrated squared density derivatives
,
1991
.
[3]
Bruce E. Hansen.
Bandwidth Selection for Nonparametric Distribution Estimation
,
2004
.
[4]
Margaret Mary McConnell,et al.
Output Fluctuations in the United States: What Has Changed Since the Early 1980s?
,
1998
.
[5]
Rodney C. Wolff,et al.
Methods for estimating a conditional distribution function
,
1999
.
[6]
Z. Q. John Lu,et al.
Nonlinear Time Series: Nonparametric and Parametric Methods
,
2004,
Technometrics.
[7]
Jianqing Fan,et al.
Local polynomial modelling and its applications
,
1994
.