Answers are grounded in this paper and cite their sources.
For the change-point model allowing at most one change X(i/n)=f(i/n)+e(i/n), where f(t)=α+θI_((t_0,1]) (t), O≤t≤1, e(1/n),…, e(n/n) are independently and identically distributed, the problem of testing and interval estimation about the change-point t_0, the magnitude of jump θ and the variance σ~2 of random error e(t), is considered.
No takes yet. Share an insight, caveat, or question.
Xiru Chen (1988) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: