.png){kind=icon}
.png){kind=icon}
Return
INFORMATION-BASED OPTIMAL SUBDATA SELECTION FOR CLUSTERWISE LINEAR REGRESSION
.png){kind=icon}
PRE
AI
DOI：10.5705/ss.202023.0302.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Mixture-of-Experts (MoE) models are commonly used when there exist distinct clusters with different relationships between the independent and dependent variables. Fitting such models for large datasets, however, is computationally virtually impossible. An attractive alternative is to use a sub data selected by maximizing the Fisher information matrix. A major challenge is that no closedform expression for the Fisher information matrix is available for such models. Focusing on clusterwise linear regression models, a subclass of MoE models, we develop a framework that overcomes this challenge. We prove that the proposed sub data selection approach is asymptotically optimal, i.e., no other method is statistically more efficient than the proposed one when the full data size is large.
Keywords:
D-optimality
information matrix
latent indicator
massive data
MLE