ExpoFit-class {RPointCloud}

The ExpoFit Class

Description

An ExpoFit object represents a robust fit to an exponential distribution, in a form that can conveniently be used as part of an Empirical Bayes approach to decompose the distributions of cycle presistence or duration for a topological data analysis performed using the TDA package.

Usage

ExpoFit(edata, resn = 200)
## S4 method for signature 'ExpoFit,missing'
plot(x, y, ...)

Arguments

edata

A numeric vector; the observed data that we think comes mainly from an exponential distribution.

resn

A numeric vector of length 1; the resolution (number of breaks) used to estimate a histogram.

x

An ExpoFit object.

y

Ignored.

...

The usual set of graphical parameters.

Value

The ExpoFit function constructs and returns an object of the ExpoFit class.

The plot method returns (invisibly) the ExpoFit object that was its first argument.

Slots

edata:

A numeric vector; the observed data that we think comes from an exponential distribution.

h0:

A histogram object produced by the hist function applied to the supplied edata.

X0:

A numeric vector containing the midpoints of the breaks in the histogram object.

pdf:

The empirical density function extracted from the histogram object.

mu:

The observed mean of the putative exponential distribution.

lambda:

The robustly estimated parameter of the exponential distribution. Originally crudely repesented by the reciprocal of the mean.

Methods

plot(x, y, lwd = 2, ...):

Produce a plot of a ExpoFit object.

Author(s)

Kevin R. Coombes <[e-mail]>

See Also

EBexpo

Examples

data(cytof)
diag <- AML10.node287.rips[["diagram"]]
persistence <- diag[, "Death"] - diag[, "Birth"]
d1 <- persistence[diag[, "dimension"] == 1]
ef <- ExpoFit(d1) # should be close to log(2)/median? 
plot(ef)