2012 Fiscal Year Final Research Report
Estimate on Regression Functions and ProbabilityDensity Functions
| Project/Area Number |
22540126
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
General mathematics (including Probability theory/Statistical mathematics)
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Renkei-kenkyūsha) |
SAKAMOTO Wataru 大阪大学, 大学院・基礎工学研究科, 准教授 (70304029)
FUJIKI Mie 同志社大学, 東アジア総合研究センター, 嘱託研究員 (70437359)
DOU Xiaoling 統計数理研究所, 新領域融合研究センター, 特任研究員 (10516868)
|
|---|
| Project Period (FY) |
2010 – 2012
|
| Keywords | 計算機統計学 / 非線形統計学 / 確率密度関数 / 2乗の積分 / kernel 関数 / U統計量 / 関数データ / 超音波データ |
|---|
| Research Abstract |
We proposed new procedures in the following two fields. First, we proposed robust estimator for simple linear regression and methods to analyze ultrasonic vocalization (USV) emitted from mice and rats. By Fourier transform ation of USV, we have functional data. We proposed methods how to eliminate errors, how to judge whether the extracted are really curves and how to classify the curves. Next, we constructed estimators of the integration of squared density function. We adopted kernel functions and U-statistics and showed that the asymptotic normality of the estimators and investigated several properties of them by computer simulations.
|
