pepcompare: Compare two fitted models

In HegemanLab/ProteinTurnover: Protein Turnover Experiment Analysis

Description

Compare two models, fitted using different parameter options.

Usage

pepcompare(pepfit1, pepfit2, sig.digits = 5)

## S3 method for class 'pepcompare'
print(x, sig.digits = x$sig.digits, ...)

Arguments

pepfit1	The first fitted model
pepfit2	The second fitted model
sig.digits	Number of significant digits to report.
x	A pepcompare object
...	Ignored, here for consistency with generic print function

Details

The models are compared by computing the change in deviance, but the sample size is unknown, so a p-value cannot be computed without assuming a given sample size.

Also, this assumes the probability distribution of deviance is chi-squared; this can be checked with the pepsim function.

Value

A list with class "pepcompare" and the following elements:

deviance	A data frame with deviance and degrees of freedom for the two models, and the difference between them.
N	The input value "N"
p.value	The p value corresponding to the input value "N"
level	The input value "level"
N.value	The sample size corresponding to the input value "level"
sig.digits	Number of significant digits to report.

Examples

  data(isodata)
 a0 <- pepfit(TimePoint, RelAb, Channel, data=isodata, Elements = list(N = 12,
    C = 45, H = 73, O = 15))
 a2 <- pepfit(TimePoint, RelAb, Channel, data=isodata, Elements = list(N = 12,
    C = 45, H = 73, O = 15), M = "one")
 pepcompare(a0, a2)

HegemanLab/ProteinTurnover documentation built on May 6, 2019, 11:50 p.m.