Abstract
I wanted to look at why bin width is important in the data analysis area for a performance study. I will be looking at various bin widths of 1.25, 2.0, 2.5, and 5.0 ns to determine optimal bin width for a performance study of newly uploaded data.
|
|
|
|
Introduction
How does changing the bin width (in ns) compare from 1.25 to 2 to 2.5 to 5.0 ns, compare for performance studies newly configured setup during our time at the Blake School in Minneapolis.
|
|
|
Procedures
I did performace studies of newly uploaded data over two days for bin widths of 1.25, 2.0, 2.5, and 5 ns.
|
|
|
Results
By using multiples of 1.25ns, the data did not truncate data points into partial bins. The plot of 2.0ns had the most non-gausian curve, because the data points were truncated into partial bins.
|
|
|
Discussions & Conclusions
By comparing the 8 different plots - (4) for the 13th of the June and the next (4) for the 14th of June - and comparing the 1.25, 2.5, and 5 ns plots looked similar to each other compared to the 2.0 ns plot, which had many high and low bins, this is becaused of the truncated data that fell into different bins.
|
|
|
|
|