Generalized estimators of the true intensity for the Illumina beadarrays, under the convolution model Rohmatul Fajriyah Institute of Statistics, TU Graz, Austria Department of Statistics, Universitas Islam Indonesia, Jogjakarta, Indonesia fajriyah@student.tugraz.at Abstract Microarray data, which come from many steps of production, have been known for containing noise. The pre-processing is implemented to reduce the noise, where the background is corrected. Prior to further analysis, many Il- lumina BeadArrays users have applied the convolution model, to estimate the true intensity value: a background corrected data, the model which has been adapted from where it was first developed on the Affymetrix platform. Several models based on the different underlying distributions and or the parameters estimation method have been proposed and applied. For instance, the exponential-gamma, the normal-gamma, and the exponential-normal con- volutions with maximum likelihood estimation, non-parametric, Bayesian and moment methods of the parameters estimation. Including a recent exponential- log normal and gamma-log normal convolutions. However none of these existing models, in the benchmarking study, performs outstandingly to others. Each of them has its own drawback. Therefore, the need to build the new model is still widely opened. i arXiv:1312.5967v1 [stat.ME] 20 Dec 2013