Calibrate Polynomial-Tail Laplace (PTL) model prdictions for LCA analysis

Share:

Description

Fits PTL models to randomly sampled pairs of the dataset, to enable prediction of PTL model parameter values based on hyperparameter d.

Usage

1
fitPTLmodel(x,nPairs=10000)

Arguments

x

Numeric data input array, standardised to range (0,1)

nPairs

Numeric value specifying the number of samplings of pairs of objects to use to obtain hyperparameter fits

Details

Evaluates parameters for PTL model fits to the distributions of feature-wise differences between each of a specified (large) number of pairs of objects represented in dataset x. Obtains subsequent model fits explaining the individual PTL parameters alpha,beta,gamma in terms of the global (Euclidean) distances between the corresponding pairs of objects.

Value

List with the following components:

alpha

Object of class lm, which can be used to predict an appropriate value of alpha in the PTL distribution corresponding to a pair of objects in the dataset with a specified global dissimilarity

beta

Object of class lm, which can be used to predict an appropriate value of alpha in the PTL distribution corresponding to a pair of objects in the dataset with a specified global dissimilarity

gamma

Object of class lm, which can be used to predict an appropriate value of alpha in the PTL distribution corresponding to a pair of objects in the dataset with a specified global dissimilarity

Author(s)

Ed Curry e.curry@imperial.ac.uk