Quantile regression approach to conditional mode estimation

In this paper, we consider estimation of the conditional mode of an outcome variable given regressors. To this end, we propose and analyze a computationally scalable estimator derived from a linear quantile regression model and develop asymptotic distributional theory for the estimator. Specifically, we find that the pointwise limiting distribution is a scale transformation of Chernoff's distribution despite the presence of regressors. In addition, we consider analytical and subsampling-based confidence intervals for the proposed estimator. We also conduct Monte Carlo simulations to assess the finite sample performance of the proposed estimator together with the analytical and subsampling confidence intervals. Finally, we apply the proposed estimator to predicting the net hourly electrical energy output using Combined Cycle Power Plant Data.

[1]  U. Grenander.  On the theory of mortality measurement , 1956 .

[2]  E. Parzen.  On Estimation of a Probability Density Function and Mode , 1962 .

[3]  H. Chernoff.  Estimation of the mode , 1964 .

[4]  C. Manski.  MAXIMUM SCORE ESTIMATION OF THE STOCHASTIC UTILITY MODEL OF CHOICE , 1975 .

[5]  T. Sager,et al.  Maximum Likelihood Estimation of Isotonic Modal Regression , 1982 .

[6]  M. R. Leadbetter,et al.  Extremes and Related Properties of Random Sequences and Processes: Springer Series in Statistics , 1983 .

[7]  P. Groeneboom.  Brownian motion with a parabolic drift and airy functions , 1989 .

[8]  Joseph P. Romano.  On weak convergence and optimality of kernel density estimates of the mode , 1988 .

[9]  D. Pollard,et al.  Cube Root Asymptotics , 1990 .

[10]  J. Horowitz.  A Smoothed Maximum Score Estimator for the Binary Response Model , 1992 .

[11]  Myoung-jae Lee,et al.  QUADRATIC MODE REGRESSION , 1993 .

[12]  Joseph P. Romano,et al.  Large Sample Confidence Regions Based on Subsamples under Minimal Assumptions , 1994 .

[13]  Jon A. Wellner,et al.  Weak Convergence and Empirical Processes: With Applications to Statistics , 1996 .

[14]  Kjell A. Doksum,et al.  On average derivative quantile regression , 1997 .

[15]  Gerard Hooghiemstra,et al.  An Extremal Limit Theorem for the Argmax Process of Brownian Motion Minus a Parabolic Drift , 1998 .

[16]  R. Koenker,et al.  Goodness of Fit and Related Inference Processes for Quantile Regression , 1999 .

[17]  J. Wellner,et al.  Computing Chernoff's Distribution , 2001 .

[18]  Arnold J Stromberg,et al.  Subsampling , 2001, Technometrics.

[19]  Dudley,et al.  Real Analysis and Probability: Measurability: Borel Isomorphism and Analytic Sets , 2002 .

[20]  Stephen Portnoy,et al.  Censored Regression Quantiles , 2003 .

[21]  Quantile Regression under Misspecification , 2004 .

[22]  C. Léger,et al.  On the bootstrap in cube root asymptotics , 2004 .

[23]  Jian Huang,et al.  ON THE BOOTSTRAP OF THE MAXIMUM SCORE ESTIMATOR , 2005 .

[24]  G. Tutz,et al.  Modelling beyond regression functions: an application of multimodal regression to speed–flow data , 2006 .

[25]  R. Koenker,et al.  Regression Quantiles , 2007 .

[26]  M. Kosorok.  Bootstrapping the Grenander estimator , 2008, 0805.2470.

[27]  Yan Yu,et al.  Single-index quantile regression , 2010, J. Multivar. Anal..

[28]  B. Sen,et al.  Inconsistency of bootstrap: The Grenander estimator , 2010, 1010.3825.

[29]  Kengo Kato,et al.  Gaussian approximation of suprema of empirical processes , 2012, 1212.6885.

[30]  Runze Li,et al.  Local modal regression , 2012, Journal of nonparametric statistics.

[31]  Hendrik P. Lopuhaa,et al.  The limit distribution of the L∞ -error of Grenander-type estimators , 2011, 1111.5934.

[32]  G. C. Kemp,et al.  Regression towards the mode , 2012 .

[33]  Kengo Kato,et al.  Valid Post-Selection Inference in High-Dimensional Approximately Sparse Quantile Regression Models , 2013, Journal of the American Statistical Association.

[34]  W. Yao,et al.  A New Regression Model: Modal Linear Regression , 2014 .

[35]  Pınar Tüfekci,et al.  Prediction of full load electrical power output of a base load operated combined cycle power plant using machine learning methods , 2014 .

[36]  Masashi Sugiyama,et al.  Modal Regression via Direct Log-Density Derivative Estimation , 2016, ICONIP.

[37]  Local M-estimation with discontinuous criterion for dependent and limited observations , 2016, 1610.02753.

[38]  Xuming He,et al.  Inference for single-index quantile regression models with profile optimization , 2016 .

[39]  L. Wasserman,et al.  Nonparametric modal regression , 2014, 1412.1716.

[40]  Haiming Zhou,et al.  Nonparametric modal regression in the presence of measurement error , 2016, 1610.08860.

[41]  S. Walker,et al.  Bayesian mode regression using mixtures of triangular densities , 2017 .

[42]  J. Krief.  Semi�?Linear Mode Regression , 2017 .

[43]  A. Yao,et al.  Non linear parametric mode regression , 2017 .

[44]  Yen-Chi Chen.  Modal regression using kernel density estimation: A review , 2017, 1710.07004.

[45]  Hendrik P. Lopuhaä,et al.  Limit Theory in Monotone Function Estimation , 2018, Statistical Science.

[46]  Alessandro Rinaldo,et al.  Distribution-Free Predictive Inference for Regression , 2016, Journal of the American Statistical Association.

[47]  Naomi S. Altman,et al.  Quantile regression , 2019, Nature Methods.

[48]  Tsuyoshi Murata,et al.  {m , 1934, ACML.

[49]  Jun Fan,et al.  A Statistical Learning Approach to Modal Regression , 2017, J. Mach. Learn. Res..

[50]  José E. Chacón,et al.  The Modal Age of Statistics , 2018, International Statistical Review.