A Brief Survey on the Choice of Parameters for: "Kernel density estimation for time series data"

Robert O'Neill, Artur Semeyutin

Research output: Contribution to journalArticle

Abstract

This paper presents an overview of the empirical performance of some of the common methods for parameter selection in the area of enhanced dynamic kernel density and distribution estimation with exponentially declining weights. It is shown that exponential weighting delivers accurate nonparametric density and quantile evaluations, without common corrections for scale and/or location in most of the financial time series considered, provided that parameters are chosen appropriately with computationally heavy Least-Squares routines. For more time-efficient numerical optimisations and/or simple kernel adaptive estimation strategies, Least-Squares routines may be re-written with exponentially weighted binned kernel estimators. This insures equally effective parameters evaluation under the different choices of kernel functional forms, though binning strategy becomes an important component of estimations. On the other hand, it is also highlighted that if the estimations target is to mine time-varying nonparametric quantiles, kernel functional forms and bandwidths may not be necessary for these evaluations. Combining exponential weights with empirical distribution estimator provides a very similar quantile performance to the kernel enhanced estimator, while parametric specifications may provide a better extreme quantiles outlook.
LanguageEnglish
JournalNorth American Journal of Economics and Finance
Early online date26 Jul 2019
Publication statusE-pub ahead of print - 26 Jul 2019

Fingerprint

Quantile
Kernel density estimation
Time series data
Evaluation
Kernel
Functional form
Kernel estimator
Least squares
Bandwidth
Time-varying
Weighting
Financial time series
Kernel density
Numerical optimization
Adaptive estimation
Empirical distribution
Estimator

Cite this

@article{3c603cd3721c495cb1655d3751114d87,
title = "A Brief Survey on the Choice of Parameters for: {"}Kernel density estimation for time series data{"}",
abstract = "This paper presents an overview of the empirical performance of some of the common methods for parameter selection in the area of enhanced dynamic kernel density and distribution estimation with exponentially declining weights. It is shown that exponential weighting delivers accurate nonparametric density and quantile evaluations, without common corrections for scale and/or location in most of the financial time series considered, provided that parameters are chosen appropriately with computationally heavy Least-Squares routines. For more time-efficient numerical optimisations and/or simple kernel adaptive estimation strategies, Least-Squares routines may be re-written with exponentially weighted binned kernel estimators. This insures equally effective parameters evaluation under the different choices of kernel functional forms, though binning strategy becomes an important component of estimations. On the other hand, it is also highlighted that if the estimations target is to mine time-varying nonparametric quantiles, kernel functional forms and bandwidths may not be necessary for these evaluations. Combining exponential weights with empirical distribution estimator provides a very similar quantile performance to the kernel enhanced estimator, while parametric specifications may provide a better extreme quantiles outlook.",
keywords = "Exponential Smoothing, Kernel Density Estimation, Binned Kernel Density and Distribution, Generalized Autoregressive Score Models, Probability Integral Transforms, Time-Varying Quantiles",
author = "Robert O'Neill and Artur Semeyutin",
year = "2019",
month = "7",
day = "26",
language = "English",
journal = "North American Journal of Economics and Finance",
issn = "1062-9408",
publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - A Brief Survey on the Choice of Parameters for

T2 - North American Journal of Economics and Finance

AU - O'Neill, Robert

AU - Semeyutin, Artur

PY - 2019/7/26

Y1 - 2019/7/26

N2 - This paper presents an overview of the empirical performance of some of the common methods for parameter selection in the area of enhanced dynamic kernel density and distribution estimation with exponentially declining weights. It is shown that exponential weighting delivers accurate nonparametric density and quantile evaluations, without common corrections for scale and/or location in most of the financial time series considered, provided that parameters are chosen appropriately with computationally heavy Least-Squares routines. For more time-efficient numerical optimisations and/or simple kernel adaptive estimation strategies, Least-Squares routines may be re-written with exponentially weighted binned kernel estimators. This insures equally effective parameters evaluation under the different choices of kernel functional forms, though binning strategy becomes an important component of estimations. On the other hand, it is also highlighted that if the estimations target is to mine time-varying nonparametric quantiles, kernel functional forms and bandwidths may not be necessary for these evaluations. Combining exponential weights with empirical distribution estimator provides a very similar quantile performance to the kernel enhanced estimator, while parametric specifications may provide a better extreme quantiles outlook.

AB - This paper presents an overview of the empirical performance of some of the common methods for parameter selection in the area of enhanced dynamic kernel density and distribution estimation with exponentially declining weights. It is shown that exponential weighting delivers accurate nonparametric density and quantile evaluations, without common corrections for scale and/or location in most of the financial time series considered, provided that parameters are chosen appropriately with computationally heavy Least-Squares routines. For more time-efficient numerical optimisations and/or simple kernel adaptive estimation strategies, Least-Squares routines may be re-written with exponentially weighted binned kernel estimators. This insures equally effective parameters evaluation under the different choices of kernel functional forms, though binning strategy becomes an important component of estimations. On the other hand, it is also highlighted that if the estimations target is to mine time-varying nonparametric quantiles, kernel functional forms and bandwidths may not be necessary for these evaluations. Combining exponential weights with empirical distribution estimator provides a very similar quantile performance to the kernel enhanced estimator, while parametric specifications may provide a better extreme quantiles outlook.

KW - Exponential Smoothing

KW - Kernel Density Estimation

KW - Binned Kernel Density and Distribution

KW - Generalized Autoregressive Score Models

KW - Probability Integral Transforms

KW - Time-Varying Quantiles

M3 - Article

JO - North American Journal of Economics and Finance

JF - North American Journal of Economics and Finance

SN - 1062-9408

ER -