Discovery of reliable causal structures in high-dimensional data
Project/Area Number |
21700302
|
Research Category |
Grant-in-Aid for Young Scientists (B)
|
Allocation Type | Single-year Grants |
Research Field |
Statistical science
|
Research Institution | Osaka University |
Principal Investigator |
SHIMIZU Shohei 大阪大学, 産業科学研究所, 助教 (10509871)
|
Project Period (FY) |
2009 – 2011
|
Project Status |
Completed (Fiscal Year 2011)
|
Budget Amount *help |
¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Fiscal Year 2011: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2010: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2009: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
|
Keywords | 統計的因果推論 / 構造方程式モデル / 独立成分分析 / 因果構造探索 / 非ガウス性 / 統計数学 / 機械学習 |
Research Abstract |
We developed several statistical methods to estimate causal networks from high-dimensional data and obtain useful causal knowledge. Specifically, we(1) developed two methods to find exogenous variables that trigger causal chains,(2) developed a direct method to estimate the entire or sub-network based on the methods for finding exogenous variables, and(3) evaluated our methods based on simulations on artificial data and real-world datasets including gene-expression data, brain imaging data and sociology data. We further made some software to perform the methods available on the internet so that many practitioners can use our methods.
|
Report
(4 results)
Research Products
(46 results)