A new statistical model helps account for excess zeros in count data
In short: Researchers have developed a Zero-Inflated Poisson regression model designed to handle datasets containing an unexpectedly high number of zero values. By treating the data as a combination of a certain zero state and a Poisson-distributed state, this method provides a more accurate way to model occurrences like manufacturing defects.
The model works by assuming a specific probability that an observation is zero, while other observations follow a standard Poisson distribution. The researchers demonstrated that these models are easy to fit using maximum likelihood estimates. While simulations suggest that confidence intervals based on likelihood ratio tests are effective, the approach's accuracy relies on having large enough samples for the estimates to be approximately normal.
Why it matters: This statistical approach allows researchers to better model processes that frequently result in zero occurrences.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Summary by Intellect · source: Research: Lambert (1992). Zero-Inflated Poisson Regression, with an Application to Defects in Manufacturing. Technometrics.
To cite this card, link to https://intellectplus.co.uk/app/science-research-a-new-statistical-model-helps-account-for-excess-zeros-in-co.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in Science · Latest · A–Z · All topics