Discrete approximations of continuous and mixed measures on a compact interval

作者：Punzo Antonio^*; Zini Alessandro

来源：Statistical Papers, 2012, 53(3): 563-575.

DOI：10.1007/s00362-011-0365-6

• Discretization
• Beta distribution
• Mixture models
• Kernel smoothing

When considering a mixed probability measure on [0, 1] with atoms in zero and one, the frontier points have to be treated (and %26quot;weighted%26quot;) differently and separately with respect to the interior points. In order to avoid the troublesome consequences of a mixed model, an easily interpretable discretization of [0, 1] on uniformly spaced atoms is here proposed. This %26quot;homogeneous support%26quot; is used to define two discrete models, a parametric and a nonparametric one. An application to real data on recovery risk of the Bank of Italy%26apos;s loans is here considered to exemplify both discretization and proposed models. Finally, a simulation study is performed to analyze the behavior of the nonparametric proposal in terms of goodness-of-fit.

• 出版日期2012-8