EM.run: Inference via EM algorithm

In jasonxu90/bdsem: Analysis of discretely observed Birth-Death-Shift processes

Description

This function runs the EM algorithm from an initial guess. Infers the coefficient vector in setting where rates depend on patient-specific covariates. The EM algorithm alternates between calling ESTEP and MSTEP until the change in observed log-likelihood changes less than a specified relative tolerance between iterations

Usage

  EM.run(betaInit, t.pat, num.patients, PATIENTDATA,
    patients.design, s1.seq, s2.seq, relTol)

Arguments

betaInit	A vector, the initial guess for coefficients beta
t.pat	A number, the observation interval length
num.patients	An integer, number of unique patients
PATIENTDATA	A matrix in the form returned by MakePatientData containing the set of observation intervals
patients.design	A design matrix in the same form as returned by PatientDesignExample
s1.seq	A vector of complex arguments evenly spaced along the unit circle
s2.seq	A vector of complex arguments evenly spaced along the unit circle
relTol	A number, the relative convergence criterion

Details

Examples are not included here due to runtime, but see vignette for usage.

Value

A list containing the log-likelihood value at convergence, the final beta estimate, and the number of iterations

jasonxu90/bdsem documentation built on May 18, 2019, 5:54 p.m.