I currently have a panel data set that contains the quarterly increment of loans initiated from more than 300 cities in China over the period from 2011Q1 to 2020Q2. I want to examine the impact of COVID-19 on lending activities. Here below is the nation-level aggregated increment of loans on a quarter base, so it is a time series. enter image description here

To avoid serious multicollinearity, I cannot use the time fixed effects and event dummy at the same time in my model. However, I do want to hold the sudden change happened in 2019Q3 to improve the model fitting. I doubt it is a structural break by just "looking", but I am unfamiliar with this technique. In my case, if it indeed is a structural break, how to detect it convincingly and adjust it? Is the structural break represented by a time-point dummy like a dummy(time >=2019Q3)?

I am also curious about a more general question. Do time fixed effects give more details than structural break? In other words, is the information told by structural break a subset of time fixed effects?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.