How can I solve non-convergence of maximum likelihood (ML) estimator in Poisson regression?

01 January 2015 1 8K Report

I am using iteratively weighted least squares algorithm to obtain ML estimates in Poisson regression. Due to the factorial term in the log-likelihood equation, I can not compute the log-likelihood value which becomes -Inf. I think it is because of some big values of the dependent variable computed from the Poisson regression.

Do you have any solution?

Martin Schmettow

I agree with Giancarlo Sadoti: a pure Poisson model is only appropriate when there are no unknown influencing variables. Have you checked for overdispersion?

When ML estimation fails, you could try MCMC estimation, the MCMCglmm library of R is a good starting point.

Badges
Science topic

More Yasin Asar's questions See All

What is a start-up value effect?

In the design of a Monte Carlo simulation, I have seen the following sentences in some of the articles: "To reduce eventual start-up value effects, we discard the first 200 observations." Can...

11 December 2014 5,639 3 View

How can I prepare virus for a TEM or SEM imaging?

I have virus (viral hemorrhagic septicemia virus) in suspension and the experiment will not involve cells. What level of TCID50 is preferred?

11 August 2024 3,115 1 View

How to learn more about SPSS and its Application?

I would like to learn more about SPSS and Its application especially in regards to data analysis. Please suggest me how I can learn more about it. Thank you so much.

11 August 2024 9,101 4 View

Can I base on reverse DNA sequences to perform alignment, convert to amino acids and GenBank submission?

I have reverse sequences (AB1 format), can I base on reverse DNA sequences to perform nucleotide alignment, convert nucleotides to amino acids and deposit the sequence in GenBank database?

11 August 2024 5,138 1 View

Baseline drift in HPLC? What causes this?

Hello, Why do i see this baseline drift when i compare my blank (black) to the sample (blue)? Any suggestions as to why this happened? Thank you!

11 August 2024 3,770 4 View

Text-Communication from the M1 Hand Area using BCI—and then there is Elon Musk?

Willett, Shenoy et al. (2021) have developed a brain computer interface (BCI) that used neural signal collected from the hand area of the motor cortex (area M1) of a paralyzed patient. The...

10 August 2024 7,180 0 View

Handling Missing Data and Building a Predictive Model with Incomplete Information ?

I am developing a predictive model for a water supply network that involves 20 influencing points. However, I only have historical data for 10 out of these 20 points. I would like to know how to...

10 August 2024 4,005 2 View

Has anyone applied Python in the field of textile engineering for data analysis, automation, or smart textiles?

I'm currently exploring the application of Python in textile engineering, specifically in areas like data analysis, process automation, and the development of smart textiles. I'm interested in...

10 August 2024 7,429 2 View

Is it possible to use the Fused Deposition Modeling (FDM) to additively manufacture interconnected porous structure generation of >100-200 micrometer?

Usually, additive manufacturing techniques like SEBM, SLS, and SLM are used for interconnected porous lattice structure generation with sizes of >100–200 micrometers. Can the Fused Deposition...

09 August 2024 7,892 0 View

How can I use the cif data obtained from rietveld refinement extracted via gsas2, for microstructural analysis using ETEX software?

09 August 2024 7,718 0 View

How to define an anisotropic material with asymmetric elastic compliance/stiffness matrix in ANSYS APDL?

I need to model an anisotropic material in which the Poisson's ratio ν_12 ≠ ν_21 and so on. Therefore, the elastic compliance matrix wouldn't be a symmetric one. In ANSYS APDL, for TB,ANEL...

09 August 2024 5,048 2 View