On the evening of December 13, 2022, the twelfth lecture of the Guanya Cross Disciplines Statement was held online. Hosted by Professor Liu Feng, Ai Chunrong, Professor of Economics and President's Chair Professor at the School of Economics and Management, The Chinese University of Hong Kong (Shenzhen), was invited to give a lecture titled "Policy Evaluation, Policy Design and Machine Learning".

Professor Liu Feng introduced the lecture topic on the application of machine learning in accounting field, such as mining annual report data using machine learning, etc. He expects that through Professor Ai Chunrong's sharing, young scholars will better combine big data tools with research problems in the field of accounting and economics to do more groundbreaking research.

Professor Ai Chunrong firstly introduced the application scenarios of econometrics, machine learning and big data in detail with the research background of interdisciplinary disciplines. Econometrics is a model-driven approach that uses models for structural extrapolation and focuses on causal inference. The advantage is that the results are interpretable, but the validity of the model specification is critical and predictions are often secondary. Machine learning (ML), on the other hand, is a data-driven approach that focuses on the selection of models in sample coverage areas to produce the "best" predictions. The results are usually uninterpretable and the validity of the model specification is not very important. Professor Ai Chunrong points out that cross-disciplinary research should ideally continue the theories and methods of econometric analysis, enrich the research variables by means of big data, and make the research more refined and clever by means of machine learning, while making the causality more relevant to reality.

Then, Professor Ai Chunrong combined specific examples to introduce the application of machine learning in policy evaluation models. Professor Ai Chunrong pointed out that although machine learning methods can provide more choices of variables, they still have to be handled differently for different scenarios in specific applications and cannot be solved simply by substitution. In addition, Professor Ai Chunrong also introduced the latest research results of machine learning in the field of policy design.

At the end of the lecture, Professor Ai Chunrong had a communication and discussion with the online teachers and students. The lecture was successfully concluded!