Rounding errors have a considerable impact on statistical inferences, especially when the data size is large and the finite normal mixture model is very important in many applied statistical problems, such as bioinformatics. In this article, we investigate the statistical impacts of rounding errors to the finite normal mixture model with a known number of components, and develop a new estimation method to obtain consistent and asymptotically normal estimates for the unknown parameters based on rounded data drawn from this kind of models.
Finite mixture normal model EM algorithm Consistent estimation Asymptotic normality
Mathematics Subject Classification (2000)
This is a preview of subscription content, log in to check access.
Mullet GM, Murray TW (1971) A new method for examining rounding error in least-squares regression computer programs. J Am Stat Assoc 66: 496–498zbMATHCrossRefGoogle Scholar
Murray A, Donald BR (1985) Estimation and hypothesis testing in finite mixture models. J R Stat Soc B 47: 67–75zbMATHGoogle Scholar
Richard JH (1985) A constrained formulation of maximum-likelihood estimation for normal mixture distributions. Ann Stat 13(2): 795–800zbMATHCrossRefGoogle Scholar
Sheppard WF (1898) On the calculation of the most probable values of frequency constants for data arranged according to equidistant divisions of a scale. Proc Lond Math Soc 29: 353–380MathSciNetzbMATHCrossRefGoogle Scholar
Tricker A (1984) The effect of rounding data sampled from the exponential distribution. J Appl Stat 11: 54–87CrossRefGoogle Scholar
Tricker A (1990a) Estimation of parameters for rounded data from non-normal distributions. J Appl Stat 17: 219–228CrossRefGoogle Scholar
Tricker A (1990b) The effect of rounding on the significance level of certain normal test statistics. J Appl Stat 17: 31–38CrossRefGoogle Scholar
Tricker A, Coates E, Okell E (1998) The effect on the R chart of precision of measurement. J Qual Technol 30: 232–239Google Scholar
Vardeman SB (2005) Sheppard’s correction for variances and the quantization noise model. IEEE Trans Instrum Meas 54: 2117–2119CrossRefGoogle Scholar