Bandwidth selection for kernel conditional density estimation
TL;DR: In this article, several bandwidth selection methods are derived ranging from fast rules-of-thumb which assume the underlying densities are known to relatively slow procedures which use the bootstrap, and a practical bandwidth selection strategy which combines the methods is proposed.
read more
About: This article is published in Computational Statistics & Data Analysis. The article was published on 28 May 2001. and is currently open access. The article focuses on the topics: Variable kernel density estimation & Conditional probability distribution.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Figures
Table 3: Bandwidth estimates for the Old Faithful Geyser data. 
Figure 9: Old Faithful Geyser data: duration of eruption plotted against waiting time to the eruption. • 
Figure 4: Estimated ISE values for each method from 50 samples. The last boxplot shows ISE values for 50 samples using the optimal values of a = 0.75 and b = 7.0. 
Figure 3: Values of b for each method from 50 samples. The dotted line shows the optimal value of b = 7.5. 
Figure 2: Values of a for each method from 50 samples. The dotted line shows the optimal value of a = 0.80. 
Figure 8: Estimated ISE values for each method from 50 samples. The last boxplot shows ISE values for 50 samples using the optimal values of a = 0.05 and b = 0.30.
Citations
•Posted Content
Estimation of conditional densities and sensitivity measures in nonlinear dynamical systems
TL;DR: Two measures of sensitivity to initial conditions in nonlinear stochastic dynamic systems are proposed, one of which relates Fisher information with initial-value sensitivity in dynamical systems and a simple method for choosing the bandwidth is proposed.
306
Using Conditional Kernel Density Estimation for Wind Power Density Forecasting
Jooyoung Jeon,James Taylor +1 more
TL;DR: This work develops an approach to producing density forecasts for the wind power generated at individual wind farms using a VARMA-GARCH model and conditional kernel density estimation, which enables a nonparametric modeling of the conditional density of wind power.
Forecasting Uncertainty in Electricity Smart Meter Data by Boosting Additive Quantile Regression
TL;DR: The empirical results demonstrate that the approachbased on quantile regression provides better forecast accuracy for disaggregated demand, while the traditional approach based on a normality assumption is a better approximation for aggregated demand.
203
Power load probability density forecasting using Gaussian process quantile regression
TL;DR: In this paper, a Gaussian Process Quantile Regression (GPQR) is proposed to handle the uncertainties in power load data in a principled manner, which can be statistically formulated.
182
A crossvalidation method for estimating conditional densities
Jianqing Fan,Tsz Ho Yim +1 more
TL;DR: In this paper, the authors extend the idea of cross-validation to choose the smoothing parameters of the double-kernel local linear regression for estimating a conditional density, which optimises the estimated conditional density function by minimising the integrated squared error.
179
References
A new look at the statistical model identification
TL;DR: In this article, a new estimate minimum information theoretical criterion estimate (MAICE) is introduced for the purpose of statistical identification, which is free from the ambiguities inherent in the application of conventional hypothesis testing procedure.
Information Theory and an Extension of the Maximum Likelihood Principle
Hirotugu Akaike
- 01 Jan 1973
TL;DR: In this paper, it is shown that the classical maximum likelihood principle can be considered to be a method of asymptotic realization of an optimum estimate with respect to a very general information theoretic criterion.
16.3K
Density estimation for statistics and data analysis
Bernard W. Silverman
- 01 Jan 1986
TL;DR: The Kernel Method for Multivariate Data: Three Important Methods and Density Estimation in Action.
Density Estimation for Statistics and Data Analysis
TL;DR: In this article, the density estimation for statistics and data analysis is presented, which is based on the Density Estimation for Statistics and Data Analysis (DESA) method. But
2.2K
•Book
Smoothing Techniques : With Implementation in S
Wolfgang Karl Härdle
- 09 Nov 2011
TL;DR: The Kernel Estimate as a Frequency Counting Curve and the Histogram as a Maximum Likelihood Estimate: Keeping the Kernel Bias the Same and Keeping the Support of the Kernel the Same.
609