Akaike (bristlecone)

Akaike Module

Namespace: Bristlecone
Assembly: Bristlecone.dll
Parent Module: ModelSelection

Functions for conducting Akaike Information Criterion (AIC).

Types

Type	Description
AkaikeWeight

Functions and values

Function or value Description


                
                  
                    
                      aic k logLikelihood

Full Usage: aic k logLikelihood

Parameters:

k : int - The number of parameters within the model in question.

logLikelihood : float - a `float` representing the minimum log-likelihood achieved for the model in question.

Returns: float

The Akaike information criterion, a standardised index of model fit quality for models that have different numbers of parameters.

k : int: The number of parameters within the model in question.
logLikelihood : float: a `float` representing the minimum log-likelihood achieved for the model in question.

Returns: float


                
                  
                    
                      aicc n k logLikelihood

Full Usage: aicc n k logLikelihood

Parameters:

n : int - The sample size

k : int - The number of parameters within the model in question

logLikelihood : float - A `float` representing the minimum log-likelihood achieved for the model in question.

Returns: float

The Akaike information criterion, corrected for small sample sizes. It represents standardised index of model fit quality for models that have different numbers of parameters.

Your model must adhere to the following assumptions: - Univariate - Linear in parameters - Normally-distributed residuals

n : int: The sample size
k : int: The number of parameters within the model in question
logLikelihood : float: A `float` representing the minimum log-likelihood achieved for the model in question.

Returns: float


                
                  
                    
                      akaikeWeights models

Full Usage: akaikeWeights models

Parameters:

models : EstimationResult seq - The input model results

Returns: (EstimationResult * AkaikeWeight) seq An (EstimationResult * float) sequence of estimation results paired to their Akaike weights.

Akaike weights for a sequence of `EstimationResult`s.

models : EstimationResult seq: The input model results

Returns: (EstimationResult * AkaikeWeight) seq: An (EstimationResult * float) sequence of estimation results paired to their Akaike weights.


                
                  
                    
                      akaikeWeightsForSet getRefCode set

Full Usage: akaikeWeightsForSet getRefCode set

Parameters:

getRefCode : 'a -> 'b - A function that gets a short reference code from a hypothesis.

set : ResultSet<'c, 'a> seq - A sequence of `ResultSet`s, within each the 1 .. many results of a particular subject * hypothesis combination.

Returns: ('c * 'b * EstimationResult * AkaikeWeight) list An `(EstimationResult * float) seq` of estimation results paired to their Akaike weights.

Akaike weights for a result set.

getRefCode : 'a -> 'b: A function that gets a short reference code from a hypothesis.
set : ResultSet<'c, 'a> seq: A sequence of `ResultSet`s, within each the 1 .. many results of a particular subject * hypothesis combination.

Returns: ('c * 'b * EstimationResult * AkaikeWeight) list: An `(EstimationResult * float) seq` of estimation results paired to their Akaike weights.

ArgumentException

Occurs when there are no observations within an estimation result.