Chinese / English
Whitney Newey


Congratulation Remarks for Gregory Chow

It’s an honor for me to be asked to offer congratulations on this occasion and I’m grateful for the opportunity.

 

Gregory has made wide range of innovative and influential contributions to economics. His “Chow Test” has for generations being the way that everybody looks to find structural change and estimating optimal controls. He developed wonderful methods of important macroeconomics. And also are precursors and foreshadow the development of dynamics of Chow models and methods and dynamic game methods, which are now so important to estimating economic structural models.

 

Along the way, he developed his interest in Chinese economy and has done wonderful things that for many years he taught very interesting and influential course for Princeton students on the Chinese economy and has written many books on the subject. It’s such an important area that economic development of China has had such a huge impact on the world in terms of make the world a better place for so many. How many many people run out of poverty resort to the Chinese economic development. And Gregory has been wonderful, and been of good influence for that and trying to point out how was that development has worked along the way as this tremendous growth, with China, this tremendous growth has happen.

 

Gregory, I know him for many years, he was my mentor and friend and host my first arrival at Princeton. So generous and kind to us and our family when we move there. They invite us to their home, show us around the city at Princeton. He also helps me learn to be a proper professor. He was such a good example, what meant to be a gentlemen, and also wonderful scholar. I’m grateful for his influence now and then on me, and which has extended down for many years. He was particular helpful in finding good Chinese food at Princeton, it is actually not so easy to do. Once you go with Gregory you know him and its fine. I’m just so grateful for his friendship, and all he has done for all these years.

 

I’m thrilled for this honor for him. The National Economics Foundation of China could not have chosen better economist to award their best Chinese economist award to. I’m thrilled to be a participant. I’m thrilled for Gregory, it’s a wonderful honor recognition for him. Congratulations to him and to the foundation for this choice.   

 

 

Congraulation Remarks for Xiaohong Chen

I’m honored to be asked to provide this congratulations for Xiaohong Chen for this award. I’m thrilled for her and thrilled for the foundation that made this award. It’s a wonderful thing. I’m grateful to offer my congratulations.

 

I met Xiaohong in an interview before her first position as an economist. And at the time she was doing interesting work on learning, which is terrific and over time her work has gone better and better. It’s just been wonderful to participate and continue to learn from her as her work has improved and becomes so important to econometrics. Her early work on learning was now very important. She developed and showed how estimate the hetero models will have formation and persistent on finance. A wonderful contribution there, she has made interesting contributions to studying the dependence over time and relationship among different independent variables. Among the things that she has done her work on non-parametric instrument variable estimation is so important and so helpful.

 

She developed results on convergence rates, on asymptotic distribution theory, and have influence for these models that has taken the results from the early identification and consistency results to just a wonderful new level and really she is the world expert in the area of non-parametric instrumental variables. That is so important in economics because it provides estimate structural models and she continues to her interest and work on series estimation, which provides a wonderful way estimate lots of different economic models. And she has really developed a theory, a wonderful way that helped us make serious progress on a lot of difficult problems.

 

It’s being wonderful to work with her. And been a co-author, I always learn things and in fact every time I talk with her, I’m grateful for the thing she sharing, because I learned new ways of think about things, and also very various way think about things, which was very helpful. Wonderful, because this foundation last for several times. I’m grateful for hospitality, thank for friendship, grateful to talk with her and look forward to bright future for her, look for what’s coming next from Xiaohong Chen. I want to commend the foundation for the wonderful choice, an economist to be awarded the best economist in China. I couldn’t think of a better person to give that to than Xiaohong Chen. She does terrific work. Congratulations to her, congratulations to Xiaohong for this award and to the foundation as well for this award.

 

 

Recommendations for future scholars

So, there is lots of interesting things that one could look at for future research suggestions for future scholars. Just such an exciting time to be working on econometrics and on empirical economics more generally. The advent of this large dataset forms the ability to analyze them to take an understand from them, what they have to say about economic behavior of agents. There is one topic that lots of interesting thing to do there. Demand analysis in large data is an interesting topic to work on. That would be great.

 

More generally, the combination of trying to answer interesting economic questions using large dataset is gonna continue to be important than very general statements. But there are lots of interesting areas that developers resolve to trying to do that. The development, for example, ways to estimate game models, new interesting areas, and analysis of networks. There has been very interesting economic development recently and some of the younger people and econometrics are leading the way in developing some of the first ways to analyze, economic networks, so that’s very exciting.

 

And more generally, the good suggestion for young scholars would be to work for interesting economic problems and trying figure out some of the very hard to work out. Game models where agents interact with each other, and dynamic programming is also challenging, it’ s a great area to work in. This is an area where large data is becoming available to be used to develop interesting important ways to predict economic outcomes, to estimate structural parameters, and those are great areas to look out. Again, more generally, you know, take some economic question that you interest or you mean to develop some methods in order to better analyze it. And care that research program is likely to be influential to economics and also do good in the world in terms of provide better macro economic tools and help with policy-makers to understand our economy better and to understand people behavior better. In the past, this is a lot to say like non-parametric instrumental variables, inference for revealed preference, estimation game models, estimation of network models, and those development can continue in anyone of these area will be better to work in.

 

 

Research Area and Major Accomplishment

Alright, some of the things that I work on, I started out doing some things on instrument variable, simultaneous equations and pretty quickly move to working on non-parametric and semi-parametric estimation. That was interesting at the time and it turns out to be really important thing in economics and econometrics, on that way could not anticipate, not good to that minute. 

 

I’ve also worked on instrument variable estimation for non-parametric models, empirical likelihood, and non-parametric panel data and other things. I worked on inference for time series and other things. And lately I’ve been interested in heterogeneity in economic models, particularly allowing for generally heterogeneity try to estimate object interesting to economist, like consumer surplus, effective nonlinear budgets and another things.

 

Economic research has been transformed over last decade or two, empirical economics has become much more important over time that has faster by increasingly variable and large macroeconomic data source, and driven by the changing interests of a lot of economic scholars. Many more model design in applied macroeconomics. Applied macro using data to answer interesting questions and now become very wide spread almost everybody does that in some form or other, which I think is a very good development for economics and is likely just gets stronger and more common. This transformation is coincided with  the economic development, and wide spread use of more sophisticated econometric methods. These methods include non-parametric and semi-parametric estimation, estimating for points and some identify structural models and estimation of causal effects. For example, in causal estimation, I just have to mention, some things that are now commonly known about years that 20 years ago people might not even heard of. This includes regression discontinuity design,  propensity scores, and of course instrumental variable estimation, all of these approaches are now parametric and semiparametric methods that are very important to study. So that’s one of the areas where there is a lot of interesting work going on right now, it’s kind of building on some of the stuffs done years ago. In particular, machine learning is top the topic now in many fields, not just econometric, but in other fields as well, computer science, other fields where data analysis are used to analyze, and thinking about how the machine learning estimate economic things that we be interested in and it is the topic that very interesting at the moment. Also those kind of massive machine learning, semiparametric and non-parametric methods. A combination of things can be used to study the economic parameters, objects like demand or choices of agents. When you have completely generalized heterogeneity that’s important to do, most of us think that people are different and the way they behavior economically and it’s important to allow for those differences to account for them. And estimating economic is interesting thing.

 

There is really a bright future in econometrics, this huge increase in applied macro has open the door for many econometric developments. When a number of people participated in, there is, you know estimation of partially identified objects which come out naturally will be a preference for scholars. The development of those econometric methods for those been very interesting and important and likely continue, for example, interesting work will include looking at machine learning methods to use on sittings and that is underway in some places I’m sure will go. There is a bright future for work and econometrics and companies that empirical macro to exciting development to come.

 



◆please indicate the source if authorized: National Economics Foundation

◆photo:National Economics Foundation