nbcf-class: Data stroge class which contain the information required for...

In kojikoji/nbcf: Change point detection for time course multi negative binomial observation

Description

This class is a data storage object for calculating change points and changing variate. You should firstly apply calculateQt and after either perfectSimulation or estimateMap.

Slots

count.mat

dgCMatrix, whose rows represent time, and columns represent variate.

t.vec

Numeric vector, whose elements represent observed time corresponding to each row of count matrix.

mean.bias.vec

Numeric vector, whose elements represent expted bias for mean value of each row.

params

List, named list of parameters, whose mebers must be "lambda", "r", "alpha" and "beta".

lpst

Numeric matrix, log probability of Ys:t

lqt

Numeric vector, log probability of Yt:n given there is change point in t-1. This will be maximum probability given k points when you give them.

lpst.list

Numeric matrix, list of lpst for each variate

lqt.list

Numeric matrix, list of lqt for each variate

bf.list

Numeric matrix, list of bayes factor for each variate. The larger values indicate the variate changed in time course.

used.vars

Character vector, variates used for change points calculation

sim.change.point.list

List, each elements contains change points simulated from perfectSimulation

map.change.point

Numeric.vector, change points simulated from estimateMap

change.variate.df

tbl, data frame of changed variates. There are columns for change points, variate and etropy of change points. Entropy is expected to be lower for moredistinct change points.

kojikoji/nbcf documentation built on June 16, 2019, 6:08 p.m.