Kernel smoothing for finite populations
β Scribed by M. C. Jones; I. S. Bradbury
- Book ID
- 104641391
- Publisher
- Springer US
- Year
- 1993
- Tongue
- English
- Weight
- 501 KB
- Volume
- 3
- Category
- Article
- ISSN
- 0960-3174
No coin nor oath required. For personal study only.
β¦ Synopsis
We identify a role for smooth curve provision in the finite population context. The performance of kernel density estimates in this scenario is explored, and they are tailored to the finite population situation especially by developing a method of data-based selection of the smoothing parameter appropriate to this problem. Simulated examples are given, including some from the particular context of permutation distributions which first motivated this investigation.
π SIMILAR VOLUMES
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 ve