lambda.network.missing {changepoints}

Function to compute the default thresholding parameter for leading singular value in the soft-impute algorithm.

Description

Function to compute the default thresholding parameter for leading singular value in the soft-impute algorithm.

Usage

lambda.network.missing(s, e, t, alpha, rho, pi_ub, p, C_lambda)

Arguments

s

An integer scalar of the starting index.

e

An integer scalar of the ending index.

t

An integer scalar of the splitting index.

alpha

A numeric scalar in (0,1) representing the desired false alarm rate.

rho

A numeric scalar of the sparsity parameter.

pi_ub

A numeric scalar of the upper bound of the missing probability.

p

An integer scalar of the dimensionality of the graphon matrix.

C_lambda

A numeric scalar of an absolute constant, which is set to be 2/3 by default.

Details

The default thresholding parameter is given in Theorem 2 of the reference.

Value

The default thresholding parameter for leading singular value in the soft-impute algorithm

References

Dubey, Xu and Yu (2021) <arxiv:2110.06450>