
1:20–2:00 | Linear models and related using Stata: More speed and more inference
Abstract:
In the last three releases, we have upgraded the most commonly used linear models in Stata, namely, areg, regress, and xtreg, fe. Some of the improvements are for speed and convenience. Some have added new options to compute valid standard errors and confidence intervals for cases in which traditional computations will underperform. And some have added estimators that leverage this technology. In this talk, I will go through these upgrades and present the theoretical results that motivate them.
Additional information:
Enrique Pinzón
StataCorp
|
2:00–2:45 | Mobile share instrumental variables and Stata application
Abstract:
The shift share instrument, also known as the “Bartik IV”, has become increasingly popular in empirical research in recent years and has made breakthroughs in theoretical research. This talk will combine classic papers to introduce the historical development and latest results of the shift share instrument, as well as the corresponding Stata practical examples.
Additional information:
Chen Qiang
Shandong University
|
2:55–3:40 | Conditional average treatment effects estimation using Stata
Abstract:
Treatment effects estimate the causal effects of a treatment on an outcome. The effect may be heterogeneous. Average treatment effects conditional on a set of variables (CATEs) help us understand heterogeneous treatment effects, and, by construction, are useful to evaluate how different treatment-assignment policies affect different groups in the population. In this talk, I will show how to use Stata's new cate command to answer questions such as the following:
Additional information:
Di Liu
StataCorp
|
3:40–4:25 | Causal inference in machine learning and Stata application
Abstract:
This presentation introduces the application of machine learning causal inference in economics, including dual machine learning estimation, inference and evaluation of conditional (heterogeneous) treatment-effect models (local linear models and interaction models), dual machine learning causal inference (DML instrumental variables, DML breakpoint design, etc.), and policy learning methods.
Additional information:
Wang Qunyong
Nankai University
|
The 2025 Chinese Stata Conference is organized by Beijing Uone Info & Tech Co., Ltd. (Uone-Tech), an official reseller of Stata in China.
View the proceedings of previous Stata Conferences and international meetings.