论文信息 - X-SOM and L-SOM: a nested approach for missing value imputation

X-SOM and L-SOM: a nested approach for missing value imputation

In this paper, a new method for the determination of missing values in temporal databases is presented. This one is based on a robust version of a nonlinear classification algorithm called Self-Organizing Maps and it consists of a combination of two classifications in order to take advantage of spatial as well as temporal dependencies of the dataset. This nested approach leads to a significant improvement of the estimation of the missing values. An application of the determination of missing values for hedge fund return database is presented.

Amaury Lendasse | Antti Sorjamaa | Bertrand Maillet | Paul Merlin

[1] Patrick Rousset,et al. Understanding and reducing variability of SOM neighbourhood structure , 2006, Neural Networks.

[2] Frank-Michael Schleif,et al. Advances in computational intelligence and learning , 2010 .

[3] M. V. Velzen,et al. Self-organizing maps , 2007 .

[4] Amaury Lendasse,et al. SOM+EOF for finding missing values , 2007, ESANN.

[5] Marie Cottrell,et al. Missing values : processing with the Kohonen algorithm , 2007, ArXiv.

[6] Sophie Midenet,et al. Self-Organising Map for Data Imputation and Correction in Surveys , 2002, Neural Computing & Applications.

[7] Tariq Samad,et al. Self–organization with partial data , 1992 .

[8] Michel Verleysen,et al. Double SOM for Long-term Time Series Prediction , 2003 .