A0960
Title: Change point detection in production frontiers
Authors: Shakeel Gavioli-Akilagun - London School of Economics (United Kingdom) [presenting]
Yining Chen - London School of Economics and Political Science (United Kingdom)
Flavio Ziegelmann - Universidade Federal do Rio Grande do Sul (Brazil)
Abstract: The focus is on the problem of retrospectively estimating locations in time at which the level of technology in an economy changes, given a sequence of time-ordered inputs and outputs. The problem is approached through the lens of stochastic frontier analysis with piecewise constant frontier functions, and an offline change point detection procedure is developed, which achieves the minimax localization rates for the problem at hand up to log factors. A simple method for constructing confidence intervals is additionally given for the unobserved change point locations. Finally, it is explained how the procedure can be modified to accommodate local changes in technology, meaning that efficiency gains are only realized for certain combinations of inputs, as well as how the procedure can accommodate serial dependence in the sequence of inputs and outputs. Simulation studies and several real data examples illustrate the practical value of the methods.
