Nadaraya. This kernel regression estimator was rst proposed by nadaraya 1964 and watson 1964. It is a linear kernel smoothing method.
The set of varying weights depends on the evaluation point x. In statistics kernel regression is a non parametric technique to estimate the conditional expectation of a random variable. In general the kernel regression estimator takes this form where k u is a kernel function.
Numeric vector with the location s at which the nadaraya watson regression estimator is to be computed.
Numeric vector x 1 x n of the x values from which together with the pertaining y values the estimate is to be computed. Numeric vector x 1 x n of the x values from which together with the pertaining y values the estimate is to be computed. Nadarayawatsonsmoother class skfda preprocessing smoothing kernel smoothers nadarayawatsonsmoother smoothing parameter none kernel function normal weights none output points none source. Note that the estimator is linear in the observations fy igand is therefore a linear smoother.
