Xiang Zhang

[rickecon]

• Good urban topic.
• Nice production function model, with analytical predictions.
• Do you do anything more than run an IV regression? You need more computational content.
• Main result is interesting, but it is just a regression result.

[Fangfangwan]

Can you introduce more policy implications to us? What might be some future applications? What changes it might bring to the society?

[zhangxiang0822]

@rickecon Hi Dr. Evans, thanks for the comment. There are three important computational components of this paper. I didn't highlight them because of the time constraint and I don't want to distract the audience from the main research question

• In the estimation of production function, I use OP and LP method, all of which requests the use of GMM method to estimate a highly non-linear system. For details, please refer to Olley and Pakes, and Levinstein and Pertin.
• The identification of structural break requires the use of time series methods and automatically identify the time and magnitude of structural breaks.
• The estimation of city TFP requires the use of Guimaraes & Portugal algorithm for estimation of a linear regression model because of the large number of fixed effects to be estimated.

Thanks

[zhangxiang0822]

@Fangfangwan 灵魂拷问啊！

[yilundai]

Very complicated yet attractive topic! Can't wait to see your future contributions!

[yilundai]

@Fangfangwan 心系社会的未来社会改革家

[Fangfangwan]

@yilundai 那必须的也不看看翔哥是什么人物！赶紧network以后孩子就有诺奖得主的推荐信了！

[Fangfangwan]

Complicated models but you explained them in an easy-to-understand way! 已经看到了你未来当一个优秀的教授的潜质！讲解的深入浅出！

[zundaxu]

The best!

[zhangxiang0822]

@zundaxu +1s!