A Law of the Iterated Logarithm for Nonparametric Regression Function Estimators
نویسندگان
چکیده
منابع مشابه
A Law of Ti~e Iterated Locarithm for Nonparametric Regression Function Estimators*
to Summary: We prove a law of the iterated logarithm for nonparametric regression function estimators using strong approximations to the two dimensional empirical process. We consider the case of Nadaraya-Watson kernel estimators and of esti-mators based on orthogonal polynomials when the marginal density of the design variable X is unknown or known.
متن کاملSequential Nonparametric Testing with the Law of the Iterated Logarithm
We propose a new algorithmic framework for sequential hypothesis testing with i.i.d. data, which includes A/B testing, nonparametric two-sample testing, and independence testing as special cases. It is novel in several ways: (a) it takes linear time and constant space to compute on the fly, (b) it has the same power guarantee (up to a small factor) as a nonsequential version of the test with th...
متن کاملOn the law of the iterated logarithm.
The law of the iterated logarithm provides a family of bounds all of the same order such that with probability one only finitely many partial sums of a sequence of independent and identically distributed random variables exceed some members of the family, while for others infinitely many do so. In the former case, the total number of such excesses has therefore a proper probability distribution...
متن کاملConvergence of nonparametric estimators for a regression function
In this paper we prove the asymptotic normality and rates of strong convergence of some types of estimators for the regression function in a xed-design regression model. We consider the Gasser-MMller estimator and the Priestley-Chao estimator (univariate and multivariate). The proofs of asymptotic normality are based on a central limit theorem from an earlier paper by the author (1996, Stochast...
متن کاملA Law of the Iterated Logarithm for General Lacunary Series
This was first proved for Bernoulli random variables by Khintchine. Salem and Zygmund [SZ2] considered the case when the Xk are replaced by functions ak cosnkx on [−π, π] and gave an upper bound ( ≤ 1) result; this was extended to the full upper and lower bound by Erdös and Gál [EG]. Takahashi [T1] extends the result of Salem and Zygmund: Consider a real measurable function f satisfying f(x + 1...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: The Annals of Statistics
سال: 1984
ISSN: 0090-5364
DOI: 10.1214/aos/1176346510