| 摘要 |
The present invention may provide a method and system for distributed regression modeling. An initial or source data set is in a partitioned form, i.e. in one or more subsets, and data and regression modules are developed separately for these data subsets. The subsets can consist of either complete data vectors or incomplete data vectors, or of a mixture of the two. The data vectors may be incomplete, for example, since they contain missing vector components. |
