Least squares estimation of a -monotone density function
| Autor: | Yong Wang, Chew-Seng Chee |
|---|---|
| Rok vydání: | 2014 |
| Předmět: |
Statistics and Probability
Mathematical optimization Applied Mathematics Explained sum of squares Least trimmed squares Generalized least squares Least squares Iteratively reweighted least squares Computational Mathematics Computational Theory and Mathematics Non-linear least squares Applied mathematics Lack-of-fit sum of squares Total least squares Mathematics |
| Zdroj: | Computational Statistics & Data Analysis. 74:209-216 |
| ISSN: | 0167-9473 |
| DOI: | 10.1016/j.csda.2014.01.007 |
| Popis: | The fact that a k-monotone density can be defined by means of a mixing distribution makes its estimation feasible within the framework of mixture models. It turns the problem naturally into estimating a mixing distribution, nonparametrically. This paper studies the least squares approach to solving this problem and presents two algorithms for computing the estimate. The resulting mixture density is hence just the least squares estimate of the k-monotone density. Through simulated and real data examples, the usefulness of the least squares density estimator is demonstrated. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect