Model Checking for Causal Effects from Economic Data
About
Correlation and causality often arise in economical data. Unlike correlation, causality can help us understand the substantially causal relationship between variables. However, studying causal effects is challenging since causality is hard to describe using statistical syntax and it is easily mixed with correlation and confounding effects. In this project, we will develop novel methods to find causal effects using model checking or testing techniques for economical data. We will also study the theories of our new methods like the consistency, asymptotic distribution of the test statistics and the power of the testing. We will evaluate the finite sample performance of our methods by simulations. Finally, we will use our proposed methods to solve economical problems, where causal effects are aimed to be found.
Required skills
Knowledge about statistical methods like causal inference and statistical theory is required.
Principal Investigator
Associate Professor, Department of Mathematics
Co-Investigator
Chair Professor, Department of Mathematics