Title: A study of stellar evolution based on structural equation modelling of stellar parameters

Authors: Sushovon Jana; Chandranath Pal

Addresses: Department of Applied Statistics, Maulana Abul Kalam Azad University of Technology, NH-12 , Simhat, Haringhata, Nadia, 741249, West Bengal, India ' Department of Statistics, University of Kalyani, Kalyani, Nadia, 741235, West Bengal, India

Abstract: A study on the sequence of changes in a star with time is an important aspect for understanding the universe. One can not observe the whole lifetime of any single star. In this paper we have considered stellar parameters at different metallicity levels of stars to accomplish stellar evolutionary studies. We have studied the changes in the relationship structures between different stellar parameters at different metallicity levels. Before doing this, we have compressed the stellar parameter space on the basis of nonlinear principal component analysis (PCA) using Neural Networks and have classified stellar parameters into two groups based on agglomerative hierarchical clustering technique around latent variables. Then we have taken into account the exploratory structural equation modelling (SEM) technique to evaluate the relationship structures between stellar parameters. Our analysis clearly highlights the changes in factorability of different stellar parameters and their associative structures over different metallicity levels.

Keywords: stellar parameter; dimension reduction; variable clustering; exploratory factors; structure equation modelling.

DOI: 10.1504/IJDATS.2024.137872

International Journal of Data Analysis Techniques and Strategies, 2024 Vol.16 No.2, pp.105 - 125

Received: 11 Mar 2023
Accepted: 11 Jan 2024
Published online: 05 Apr 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article