Skip to contents

Scans a grid of candidate partition points with increments of size 1, as described in the manuscript, and selects the point minimising the pooled mean squared error, subject to a minimum number of observations on each side.

Usage

find_partition_point(t, y, min_obs = 12L, criterion = "mse", beta_fixed = 1.5)

Arguments

t, y

Time index and cumulative counts of the open window.

min_obs

Minimum number of observations required on each side of the candidate split (default 12, as in Simulation Scenarios 1 and 3 of the manuscript; use 15 to match Scenario 2).

criterion, beta_fixed

Passed to fit_ensemble().

Value

A list with grid (candidate points searched), mse_profile (MSE at each grid point), d_hat (selected partition point), and mse_split (MSE at d_hat).