BICmodel: Bayesian Information Criterion (BIC)

In genomaths/MethylIT: Methylation Analysis Based on Signal Detection

Bayesian Information Criterion (BIC)

Description

this function permits the estimation of the BIC for models for which the function 'BIC' from 'stats' packages does not work.

Usage

BICmodel(model = NULL, residuals = NULL, np = NULL)

Arguments

model	if provided, it is an R object from where the residuals and model parameters can be retrieved using resid(model) and coef(model), respectively.
residuals	if provided, it is numerical vector with the residuals: residuals = observe.values - predicted.values, where predicted values are estimated from the model. If the parameter 'model' is not provided, then this parameter must be provided.
np	number of model parameters. If the parameter 'model' is not provided, then 'np' and 'residuals' must be provided.

Details

if for a given model 'm' BIC(m) works, then BICmodel(m) = BIC(m).

Value

BIC numerical value

Examples

## Build a set of point holding an exponential decay
## and adding random noise.
set.seed(77)
x = runif(100, 1, 5)
y = 2 * exp(-0.5 * x) + runif(100, 0, 0.1)
plot(x, y)

## Non-linear regression
nlm <- nls(Y ~ a * exp( b * X), data = data.frame(X = x, Y = y),
            start = list( a = 1.5, b = -0.7),
            control = nls.control(maxiter = 10^4, tol = 1e-05),
            algorithm = 'port')

## The estimations of Akaike information criteria given by BIC' function
## from stats' R package and from 'AICmodel' function are equals.
BICmodel(nlm) == BIC(nlm)

## Now, using residuals from the fitted model:
res = y - coef(nlm)[1] * exp(coef(nlm)[2] * x)

BICmodel(residuals = res, np = 2) == BIC(nlm)

genomaths/MethylIT documentation built on Feb. 3, 2024, 1:24 a.m.