Ja Kobi Cockerham

Ja Kobi Cockerham

Biography

Ja Kobi is a rising sophomore studying physics and mathematics at the illustrious
Howard University. Currently, he is conducting research at MIT’s Laboratory for Nuclear
Science through the MIT Summer Research Program, where he works on the precise
measurement of the strong coupling constant by reproducing the Z-boson in Drell-Yan events.
His infatuation with physics began when he watched Gravity Falls as a child, a show that made
the universe feel like something worth decoding. That early wonder sharpened into a long-term
goal: pursuing graduate research in string theory and quantum field theory, chasing the deeper
structures beneath the physics he studies today. Outside the lab, Ja Kobi volunteers with the
YMCA, helps special needs children achieve their physical education goals, and studied abroad
in Ghana. He spends his free time weightlifting, watching films, and reading Dostoyevsky,
carrying the same discipline and curiosity into every room he enters.

Smoothing Statistical and Systematic Uncertainty Bands in MiNNLO Predictions
for Precision αs Extraction via Drell-Yan Z-boson Production
Ja Kobi D. Cockerham1, David Walter2, Luca Lavezzo2 and Christoph M.E. Paus2
1Department of Physics and Astronomy, Howard University
2Department of Physics, Massachusetts Institute of Technology

Precision measurements of fundamental particle interactions rely on comparing experimental
data to theoretical predictions. However, finite Monte Carlo statistics used to estimate these
predictions’ uncertainties can introduce noisy, non-physical fluctuations across measurement
ranges. One vital input to such measurements, the strong coupling constant, is extracted
from Drell-Yan events, yet current uncertainty bands derived from MiNNLO Monte Carlo
predictions often carry bin-to-bin statistical noise, complicating their use in precise extractions.
This project develops a procedure to characterize and smooth these statistical and systematic
uncertainty bands across key kinematic variables in simulated collision data, using Pythonbased
statistical and histogram analysis tools to isolate genuine kinematic-dependent trends
from random fluctuation. Initial application of this procedure to a subset of simulated events
demonstrates its ability to distinguish structured trends from noise, particularly across the
transverse momentum of the Z boson. Full application of this method across the complete
dataset is expected to yield a quantified reduction in uncertainty band variance, directly
improving the precision with which the coupling constant can be extracted. This approach
offers a generalizable framework for incorporating smoothed uncertainty estimates into future
precision measurements, with relevance to ongoing efforts across particle physics to refine
fundamental constants with greater confidence.