Local variance estimation for uncensored and censored observations
Author(s)
Bibliographic Information
Local variance estimation for uncensored and censored observations
（Research）
Springer Vieweg, c2013
Available at / 1 libraries
-
No Libraries matched.
- Remove all filters.
Note
Originally presented as the author's thesis (doctoral) -- Universität Stuttgart, 2012
Description and Table of Contents
Description
Paola Gloria Ferrario develops and investigates several methods of nonparametric local variance estimation. The first two methods use regression estimations (plug-in), achieving least squares estimates as well as local averaging estimates (partitioning or kernel type). Furthermore, the author uses a partitioning method for the estimation of the local variance based on first and second nearest neighbors (instead of regression estimation). Approaching specific problems of application fields, all the results are extended and generalised to the case where only censored observations are available. Further, simulations have been executed comparing the performance of two different estimators (R-Code available!). As a possible application of the given theory the author proposes a survival analysis of patients who are treated for a specific illness.
Table of Contents
Least Squares Estimation of the Local Variance via Plug-In.- Local Averaging Estimation of the Local Variance via Plug-In.- Partitioning Estimation of the Local Variance via Nearest Neighbors.- Estimation of the Local Variance under Censored Observations.
by "Nielsen BookData"
