𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Smoothing for Discrete Kernels in Discrimination

✍ Scribed by Dr. Gerhard Tutz

Publisher
John Wiley and Sons
Year
2007
Tongue
English
Weight
514 KB
Volume
30
Category
Article
ISSN
0323-3847

No coin nor oath required. For personal study only.

✦ Synopsis

In multivariate discrimination by the discrete kernel method the allocation rule is Bayrisk consistent if the smoothing parameter is choeen by maximizstion of the leaving-one-out nonerror rate. It is shown that oonaistency still holda if the leeving-one-out nonerror rate is replaced by a smoothed version. Thue a om-validetory criterion is given which ~ecurea consistency and really can be used in practice.

πŸ“œ SIMILAR VOLUMES

Asymptotic Normality for Density Kernel
Asymptotic Normality for Density Kernel Estimators in Discrete and Continuous Time
✍ Denis Bosq; Florence MerlevΓ¨de; Magda Peligrad πŸ“‚ Article πŸ“… 1999 πŸ› Elsevier Science 🌐 English βš– 153 KB

In this paper, we build a central limit theorem for triangular arrays of sequences which satisfy a mild mixing condition. This result allows us to study asymptotic normality of density kernel estimators for some classes of continuous and discrete time processes.

An efficient fixed-lag smoothing algorit
An efficient fixed-lag smoothing algorithm for discrete linear systems
✍ Gerald J. Bierman πŸ“‚ Article πŸ“… 1974 πŸ› Elsevier Science 🌐 English βš– 446 KB

The modified Bryson-Frazier fixed interval smoothing algorithm [6], is an addendem to the Kalman filterβ€’ This algorithm when applied to the problem of fixedlag smoothing is computationally more efficient than the algorithms recently reported in refs. [1][2][3]. Features of the algorithm are ease of

Results of von Neumann analyses for repr
Results of von Neumann analyses for reproducing kernel semi-discretizations
✍ Mark A. Christon; Thomas E. Voth πŸ“‚ Article πŸ“… 2000 πŸ› John Wiley and Sons 🌐 English βš– 178 KB

The reproducing kernel particle method (RKPM) has many attractive properties that make it ideal for treating a broad class of physical problems. RKPM may be implemented in a 'mesh-full' or a 'mesh-free' manner and provides the ability to tune the method, via the selection of a window function and it

Covariance Adjustments in Discrimination
Covariance Adjustments in Discrimination of Mixed Discrete and Continuous Variables
✍ Chi-Ying Leung πŸ“‚ Article πŸ“… 1999 πŸ› Elsevier Science 🌐 English βš– 126 KB

Sufficient conditions are given to ensure a better performance of the plug-in version of the covariates adjusted location linear discriminant function in an asymptotic comparison of the overall expected error rate. Our findings generalize several earlier results on discriminant function with covaria