Beta_ID: Beta Prior Distribution Identifier

Description Usage Arguments Details Value Author(s) Examples

Description

Uses the subject matter researcher's knowledge to generate a corresponding Beta prior distribution.

Usage

1
Beta_ID(Low, High, Cover = NULL)

Arguments

Low

researcher's LOWEST plausible value for the parameter.

High

researcher's HIGHEST plausible value for the parameter.

Cover

researcher's suggested coverage for the Low and High values provided.

Details

Uses non-linear minimization to provide an optimal Beta prior distribution.

Value

Provides graphical as well as full textual description of a suitable Beta distribution for researcers based on their knowledge about how High or Low the parameter has been found in the liliter?ature. Also, helps researcher to revise their prior by issuing various messages.

Author(s)

Reza Norouzian <rnorouzian@gmail.com>

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
# Suppose a researcher needs a Beta prior for a parameter that in his/her
# view can't be less than 2% and more than 90%. The researcher believes
# these two limit values cover 90% of all possible values that the parameter
# of interest to him/her can take:


 Beta_ID (Low = "2%", High = "80%", Cover = '90%')



# User can also use any value that is between 0 and 1 for all arguments
# without using percentage sign for some or all arguments:


Beta_ID (Low = .02, High = .8, Cover = "90%")


Beta_ID (Low = .02, High = .8, Cover = 90)

rnorouzian/BayesianforL2 documentation built on May 29, 2019, 8:37 a.m.