Efficient B-spline imputation methods in functional structural equation model with missing data
-
Hao Cheng [1]
-
[1] Minzu University of China
Minzu University of China
China
-
- Localización: Test: An Official Journal of the Spanish Society of Statistics and Operations Research, ISSN-e 1863-8260, ISSN 1133-0686, Vol. 34, Nº. 4, 2025, págs. 1036-1079
- Idioma: inglés
- Texto completo no disponible (Saber más ...)
-
Resumen
Missing data has been one of the commonly seen phenomena in various applications. The paper proposes two kinds of efficient B-spline imputation methods of handling missing data problems based on estimation equations projection and its extension combined with inverse probability weighting for functional structural equation model, respectively. Each efficient B-spline imputation method allows functional loading coefficients and functional path coefficients with missing data to be estimated at different quantiles. Through simulation investigations, both bootstrap and bag of little bootstraps are introduced to illustrate the performances of our proposed efficient B-spline imputation methods in functional structural equation model. Finally, both our proposed models and methods are applied to a real data example on Enterprise Innovation Growth Index.
