Work

The application of adaptive group LASSO imputation method with missing values in personal income compositional data

Year: 2024

Type: article

Abstract: From social and economic perspectives, compositional data represent the proportions of various components within a whole, carrying non-negative values and providing only relative information. However, in many circumstances, there are often a significant number of missing values in datasets. Due to the complexity caused by these missing values, traditional estimation methods are ineffective. In this paper, an adaptive group LASSO-based imputation method is proposed for compositional data, consolidating the advantages of group LASSO and adaptive LASSO analysis techniques. Considering the impact of outliers on the accuracy of estimation, both simulation and case analysis are conducted to compare the proposed algorithm against four existing methods. The experimental results demonstrate that the proposed adaptive group LASSO method produces a better imputation performance at comparable missing rates.

Source: Journal Of Big Data

DOI: 10.1186/s40537-024-01009-1

Authors: Ying Tian, Majid Khan Majahar Ali, Lili Wu

Institutions: Universiti Sains Malaysia, Taiyuan Institute of Technology, Xinzhou Teachers University

Language: English

Topic: Statistical Methods and Bayesian Inference

Subfield: Statistics and Probability

Field: Mathematics

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $2,090

Funders: National Bureau of Statistics of China, Universiti Sains Malaysia, Division of Mathematical Sciences