Nonparametric Kernel Method to Hedge Downside Risk

AbstractCredibility weighting is helpful in many insurance applications where sparse data crave information from other sources of data. In this paper we aim at estimating a hazard curve using the nonparametric kernel method, where a credibility weighting principle is used locally, so that areas of sparse data for one subgroup can be alleviated by available information from other subgroups. The credibility estimator is found through a Hilbert space projection formulation of Buhlmann-Straub's credibility approach.

Download Full-text

Testing independence by nonparametric kernel method

Statistics & Probability Letters ◽

10.1016/s0167-7152(96)00183-6 ◽

1997 ◽

Vol 34 (2) ◽

pp. 201-210 ◽

Cited By ~ 16

Author(s):

Ibrahim A. Ahmad ◽

Qi Li

Keyword(s):

Kernel Method ◽

Testing Independence ◽

Nonparametric Kernel ◽

Nonparametric Kernel Method

Download Full-text

Asymptotic Properties for Weighted Averages of Longitudinal Dependent Data

International Journal of Statistics and Probability ◽

10.5539/ijsp.v7n6p100 ◽

2018 ◽

Vol 7 (6) ◽

pp. 100

Author(s):

Brahima Soro ◽

Ouagnina Hili ◽

Youssouf Diagana

Keyword(s):

Longitudinal Data ◽

Covariance Function ◽

Kernel Method ◽

Asymptotic Properties ◽

Dependent Data ◽

Independent Data ◽

Function Estimation ◽

Mixing Conditions ◽

Weighted Averages ◽

Nonparametric Kernel

This paper presents a set of normality general results for kernel weighted averages. We extend existing literature for independent data (Yao, 2007) to stationary dependent longitudinal data. The asymptotic properties of proposed weighted averages are investigate under α-mixing conditions. These results are useful for covariance function estimation based on nonparametric kernel method.

Download Full-text

Nonparametric instrumental variables for identification of block-oriented systems

International Journal of Applied Mathematics and Computer Science ◽

10.2478/amcs-2013-0040 ◽

2013 ◽

Vol 23 (3) ◽

pp. 521-537 ◽

Cited By ~ 2

Author(s):

Grzegorz Mzyk

Keyword(s):

Rate Of Convergence ◽

Instrumental Variables ◽

Least Squares Method ◽

Kernel Method ◽

Estimation Error ◽

Identification Algorithm ◽

Correlated Noise ◽

The Matrix ◽

Nonparametric Kernel ◽

Special Case

Abstract A combined, parametric-nonparametric identification algorithm for a special case of NARMAX systems is proposed. The parameters of individual blocks are aggregated in one matrix (including mixed products of parameters). The matrix is estimated by an instrumental variables technique with the instruments generated by a nonparametric kernel method. Finally, the result is decomposed to obtain parameters of the system elements. The consistency of the proposed estimate is proved and the rate of convergence is analyzed. Also, the form of optimal instrumental variables is established and the method of their approximate generation is proposed. The idea of nonparametric generation of instrumental variables guarantees that the I.V. estimate is well defined, improves the behaviour of the least-squares method and allows reducing the estimation error. The method is simple in implementation and robust to the correlated noise.

Download Full-text