Distribution-free consistency of kernel
β
Andrzej S. Kozek; MirosΕaw Pawlak
π
Article
π
2002
π
Elsevier Science
π
English
β 124 KB
We prove that in the case of independent and identically distributed random vectors (X i ; Y i ) a class of kernel type M-estimators is universally and strongly consistent for conditional M-functionals. The term universal means that the strong consistency holds for all joint probability distribution