An Enhanced Methodology for Estimating Disease Infection Rates in Plants Using Unequal Group Size Testing: A Statistical Approach

• Author(s): Sirengo John Luca
• Paper ID: 1706641
• Page: 76-82
• Published Date: 11-12-2024
• Published In: Iconic Research And Engineering Journals
• Publisher: IRE Journals
• e-ISSN: 2456-8880
• Volume/Issue: Volume 8 Issue 6 December-2024

Abstract

Group testing has emerged as a crucial methodology for efficient disease detection in plant populations, particularly when dealing with unequal group sizes. This study presents an enhanced statistical framework for estimating disease infection rates in plants using unequal group size testing, building upon Dorfman's (1943) foundational work. The research develops and analyzes a maximum likelihood estimator for unequal group sizes, expressed as ?? = 1-(1-x_i/n_i)^(1/k_i), where x_i represents positive groups, n_i the number of groups, and k_i the group size. Through extensive simulation studies, we demonstrate that this estimator exhibits minimal bias (< 0.001) for infection rates below 0.30 and achieves significant reductions in mean square error (67% reduction compared to one-at-a-time testing for ? = 0.10 and k = 15). Our findings reveal remarkable improvements in testing efficiency, with asymptotic relative efficiency values ranging from 34.0 to 80.27 across different infection rates and group sizes. The optimal group size analysis indicates that larger groups (k > 10) are most efficient for infection rates below 0.15, leading to a 78% reduction in testing costs while maintaining statistical power above 0.90. Additionally, the study demonstrates that moderate variation in group sizes (CV ? 0.3) has minimal impact on efficiency, making the methodology practically applicable in real-world scenarios where equal group sizes may not be feasible.

Keywords

Unequal Group Size, Plant Disease Detection, Maximum Likelihood Estimation, Asymptotic Relative Efficiency, Optimal Group Size

Citations

IRE Journals:
Sirengo John Luca "An Enhanced Methodology for Estimating Disease Infection Rates in Plants Using Unequal Group Size Testing: A Statistical Approach" Iconic Research And Engineering Journals Volume 8 Issue 6 2024 Page 76-82

IEEE:
Sirengo John Luca "An Enhanced Methodology for Estimating Disease Infection Rates in Plants Using Unequal Group Size Testing: A Statistical Approach" Iconic Research And Engineering Journals, 8(6)