EXPLORATION OF DISTRIBUTION’S PARAMETERS AND CONFIDENCE INTERVAL

The pattern formed from a risk data can be studied and represented become a certain distribution. Each distribution has a component that can affect the formed pattern, namely parameters. Parameter is a value that can describe the population. The unknown parameter can be estimated by parameter est...

Full description

Saved in:
Bibliographic Details
Main Author: Ismail Walid, M.
Format: Theses
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/50102
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
id id-itb.:50102
spelling id-itb.:501022020-09-22T13:34:36ZEXPLORATION OF DISTRIBUTION’S PARAMETERS AND CONFIDENCE INTERVAL Ismail Walid, M. Indonesia Theses data pattern, parameter, point estimation, confidence interval, risk models, bivariate Copulas INSTITUT TEKNOLOGI BANDUNG https://digilib.itb.ac.id/gdl/view/50102 The pattern formed from a risk data can be studied and represented become a certain distribution. Each distribution has a component that can affect the formed pattern, namely parameters. Parameter is a value that can describe the population. The unknown parameter can be estimated by parameter estimation which can be either a point estimation or an interval estimation. The method that can be used for point estimation is the maximum likelihood estimation method and the Bayesian estimation method. The maximum likelihood estimation method can be used for several distribution parameters such as the ?, ?2, p, or the parameters for the aggregate risk models. Meanwhile, the Bayesian estimation method is used to estimate the parameters treated as random variables. The point estimation is considered to be less representative of the parameter value, because it is only a single value, so an interval estimation is required. One of the more popularly interval estimation is the confidence interval. The confidence interval of a parameter is the point estimate value added (subtracted) by a statistic which multiplied by the standard error. In addition to single risk data, parameter estimation can also be determined for paired (bivariate) risk data. For bivariate paired data from different distributions, bivariate Copula can be used as an alternative to obtain a joint distribution function. From the Copula, the parameter values can also be estimated using the maximum likelihood estimation method. From the point estimation for the Copula parameter, the confidence interval can be determined. The criteria for a good confidence interval can be seen from the length of the interval and its probability (level of confidence). text
institution Institut Teknologi Bandung
building Institut Teknologi Bandung Library
continent Asia
country Indonesia
Indonesia
content_provider Institut Teknologi Bandung
collection Digital ITB
language Indonesia
description The pattern formed from a risk data can be studied and represented become a certain distribution. Each distribution has a component that can affect the formed pattern, namely parameters. Parameter is a value that can describe the population. The unknown parameter can be estimated by parameter estimation which can be either a point estimation or an interval estimation. The method that can be used for point estimation is the maximum likelihood estimation method and the Bayesian estimation method. The maximum likelihood estimation method can be used for several distribution parameters such as the ?, ?2, p, or the parameters for the aggregate risk models. Meanwhile, the Bayesian estimation method is used to estimate the parameters treated as random variables. The point estimation is considered to be less representative of the parameter value, because it is only a single value, so an interval estimation is required. One of the more popularly interval estimation is the confidence interval. The confidence interval of a parameter is the point estimate value added (subtracted) by a statistic which multiplied by the standard error. In addition to single risk data, parameter estimation can also be determined for paired (bivariate) risk data. For bivariate paired data from different distributions, bivariate Copula can be used as an alternative to obtain a joint distribution function. From the Copula, the parameter values can also be estimated using the maximum likelihood estimation method. From the point estimation for the Copula parameter, the confidence interval can be determined. The criteria for a good confidence interval can be seen from the length of the interval and its probability (level of confidence).
format Theses
author Ismail Walid, M.
spellingShingle Ismail Walid, M.
EXPLORATION OF DISTRIBUTION’S PARAMETERS AND CONFIDENCE INTERVAL
author_facet Ismail Walid, M.
author_sort Ismail Walid, M.
title EXPLORATION OF DISTRIBUTION’S PARAMETERS AND CONFIDENCE INTERVAL
title_short EXPLORATION OF DISTRIBUTION’S PARAMETERS AND CONFIDENCE INTERVAL
title_full EXPLORATION OF DISTRIBUTION’S PARAMETERS AND CONFIDENCE INTERVAL
title_fullStr EXPLORATION OF DISTRIBUTION’S PARAMETERS AND CONFIDENCE INTERVAL
title_full_unstemmed EXPLORATION OF DISTRIBUTION’S PARAMETERS AND CONFIDENCE INTERVAL
title_sort exploration of distribution’s parameters and confidence interval
url https://digilib.itb.ac.id/gdl/view/50102
_version_ 1822272257910636544