Abstract
Here we adopt the method of estimation for the derivatives of a probability density function based on wavelets discussed in Prakasa Rao (1996) to the case of negatively associated random variables. An upper bound on Lp-loss for the resulting estimator is given which extends such a result for the integrated mean square error (IMSE) given in Prakasa Rao (1996). Also, considering the case of derivative of order zero, the results given by Kerkyacharian and Picard (1992), Tribouley (1995) and Leblanc (1996) are obtained as special cases.
Original language | English |
---|---|
Pages (from-to) | 453-463 |
Number of pages | 11 |
Journal | Journal of Statistical Theory and Practice |
Volume | 2 |
Issue number | 3 |
DOIs | |
Publication status | Published - 2008 |
Externally published | Yes |
Keywords
- Besov space
- multiresolution analysis
- negative dependence
- nonparametric estimation
- wavelets