Advanced

Isotonic regression in deterministic and random design settings

Jönsson, Viktor (2016) MASM01 20161
Mathematical Statistics
Abstract (Swedish)
This thesis will treat the subject of constrained statistical inference
and will have its focus on isotonic regression, which is the problem of
estimating functions that are assumed to be monotone.
A characterisation of isotonic regression and a solution to this problem
will be given. The PAVA algorithm to compute the isotonic regres-
sion estimator will be introduced along with asymptotic distribution
results for this. The aim is to investigate the properties of the esti-
mator when the observation points are random dependent variables
which also depend on the unknown function itself.
Please use this url to cite or link to this publication:
author
Jönsson, Viktor
supervisor
organization
course
MASM01 20161
year
type
H2 - Master's Degree (Two Years)
subject
language
English
id
8568359
date added to LUP
2016-01-26 11:42:04
date last changed
2016-01-26 11:42:04
@misc{8568359,
  abstract     = {This thesis will treat the subject of constrained statistical inference
and will have its focus on isotonic regression, which is the problem of
estimating functions that are assumed to be monotone.
A characterisation of isotonic regression and a solution to this problem
will be given. The PAVA algorithm to compute the isotonic regres-
sion estimator will be introduced along with asymptotic distribution
results for this. The aim is to investigate the properties of the esti-
mator when the observation points are random dependent variables
which also depend on the unknown function itself.},
  author       = {Jönsson, Viktor},
  language     = {eng},
  note         = {Student Paper},
  title        = {Isotonic regression in deterministic and random design settings},
  year         = {2016},
}