estimateNoise: Estimate and model noise in signal

In ppernot/FitOCTlib: Stan-based functions to fit OCT signals

Description

Estimate and model noise in signal

Usage

estimateNoise(x, y, df = 15, maxRate = 10000)

Arguments

x	a numeric vector
y	a numeric vector of responses
df	smoothing factor for smooth.splines
maxRate	max. value of rate parameter

Details

Function which proceeds in two steps:

1. get a set of residuals R using a smoothing splines model

2. estimate the x-dependent standard deviation of the residuals

by bayesian inference: R(x) ~ normal(0,uy(x)); uy(x) = theta[1]*exp(-x/theta[2]) assuming a Poisson-type noise.

Value

A list containing

fit

a stanfit object containg the results of the fit

method

choice of optimization method

theta

a vector of optimal parameters

uy

a vector of estimated uncertainty values for y

ySmooth

a vector of values for the smoother curve

Author(s)

Pascal PERNOT

ppernot/FitOCTlib documentation built on April 11, 2020, 1:55 a.m.