Activity Number:
|
584
- Advances in Semi- and Nonparametric Statistical Analysis
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 1, 2018 : 2:00 PM to 3:50 PM
|
Sponsor:
|
IMS
|
Abstract #330092
|
Presentation
|
Title:
|
Generalized Semiparametric Approach to One-Way Analysis of Variance
|
Author(s):
|
Chathurangi Pathiravsan* and Bhaskar Bhattacharya
|
Companies:
|
Southern Illinois University Carbondale and Southern Illinois University Carbondale
|
Keywords:
|
Semi-parametric Approach;
One-Way Analysis of Variance;
Minimum Discriminant Information Adjustment
|
Abstract:
|
The Analysis of Variance (ANOVA) is an extremely important tool for analysis of data in many fields. In one-way ANOVA, hypothesis testing allows us to test the equality of means of different levels of a factor simultaneously using variance. To relax the normal assumption in one-way ANOVA, the recent studies have considered exponential distortion or tilt of a reference distribution. The reason for the exponential distortion was not investigated before thus the reason was examined with the help of duality and minimum discriminant information adjustment (MDIA) concepts. A new direction of generalized semi-parametric approach in terms of MDIA is proposed to compare means of more than two groups. This generalized approach relaxes all the assumptions in one-way ANOVA and it can be applied to any type of distribution. The performance of the proposed approach is demonstrated on simulated data examples, and is compared with the existing techniques for one-way ANOVA.
|
Authors who are presenting talks have a * after their name.