Answers are grounded in this paper and cite their sources.
If ( X 1 , Y 1 ), …, ( X n ,Y n ) is a sequence of independent identically distributed R d × R ‐valued random vectors then Nadaraya (1964) and Watson (1964) proposed to estimate the regression function m(x) = ϵ {Y 1 |X 1 = x{ by where K is a known density and { h n } is a sequence of positive numbers satisfying certain properties. In this paper a variety of conditions are given for the strong convergence to 0 of ess X sup| m n ( X )‐ m ( X )| (here X is independent of the data and distributed as X 1 ). The theorems are valid for all distributions of X 1 and for all sequences { h n } satisfying h n → 0 and nh /log n→0.
No takes yet. Share an insight, caveat, or question.
Luc Devroye (1978) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: