Interval Estimation of a Normal Process Mean from Rounded Data
Standard statistical methods are based on an implicit assumption that numerical data are exact. But in truth, all real data are rounded to some smallest unit of measure related to the precision of the device used to produce them. When the degree of rounding is severe, ignoring the rounding produces statistical methods with operating characteristics far from nominal. We discuss the interval estimation of the parameter μ when rounded data come from the N(μ,σ2) distribution.
This article is published as Lee, Chiang-Sheng, and Stephen B. Vardeman. "Interval estimation of a normal process mean from rounded data." Journal of Quality Technology 33, no. 3 (2001): 335-348. DOI: 10.1080/00224065.2001.11980083. Posted with permission.