EVSnoopingCV: Critical values based on extreme value approximation

In kolesarm/BWSnooping: Confidence invervals robust to bandwidth snooping

Description

Calculate one- and two-sided critical values based on extreme value approximation to the limiting distribution as the ratio of maximum to minimum bandwidth diverges to infinity. Not recommended to use in practice (the critical values provided by SnoopingCV are more accurate), only for illustration.

Usage

EVSnoopingCV(bwratio, kernel, boundary, order, onesided = FALSE,
  alpha = 0.05)

Arguments

bwratio	ratio of maximum to minimum bandwidth, number greater than 1
kernel	Either one of "uniform", "triangular", or "epanechnikov", or else an (equivalent) kernel function supported on [-1,1].
boundary	Logical specifying whether regression is in the interior or on the boundary, and an integer specifying order of local polynomial. If kernel is "uniform", "triangular", or "epanechnikov", the appropriate boundary or interior equivalent kernel is used. If kernel is a function, these options are ignored.
order	Logical specifying whether regression is in the interior or on the boundary, and an integer specifying order of local polynomial. If kernel is "uniform", "triangular", or "epanechnikov", the appropriate boundary or interior equivalent kernel is used. If kernel is a function, these options are ignored.
onesided	Logical specifying whether the critical value corresponds to a one-sided confidence interval.
alpha	number specifying confidence level, 0.05 by default.

kolesarm/BWSnooping documentation built on May 20, 2019, 12:54 p.m.