Tag Archives: econometrics

Heckman's talk on inequity

Today, I was so luck to have enjoyed Heckman's speech. As usual, the information is attached here:

MOVE Distinguished Visitor Lecture
Speaker: James Heckman
Date: March 24, 2011

Heckman is well-known for the development of theoretical and empirical models of human development and lifecycle skill formation, with a special emphasis on the economics of early childhood education.

To be honest, I know Heckman because of his famous contribution to theoretical econometrics - actually, I just learned Heckman two-step estimation this month... Then he came here, and I got the chance to see the "real" version. Out of my expectation, he is such a good speaker! I concentrated on his speech for two continuous hours without a second to relax. OMG... I thought he was a technical guy - from the typical impression he should not care about anything else but math... However, I was totally wrong! He also works a lot on applications and empirical works which reflects the responsibility of a real economist. Like today, he talked about inequity. Although he was going to "sell" his academic idea, it was so good and so intuitive that can really convince you what is going on in the real world. So fresh, and comes to you with tons of inspirations.  Nice~

Tomorrow the more academic-targeted workshop will take place in UAB. Last time when I went to MOVE's workshop, there were only 20 people in a very big lecture hall and some of them were even sleeping... This time, from today's observation, I think there will be over 100 people tomorrow, and a big proportion of them will be brilliant researchers. It should be impressive. The basic info is attached here:

ICREA-MOVE Conference on Family Economics

Program Committee:Pierre-Andre Chiappori, Christopher Flinn, Jeremy Greenwood, Nezih Guner and James Heckman
Local organization: Nezih Guner
Date: March 25-26, 2011
Venue
: Campus de Bellaterra-UAB

It is so nice to enjoy this kind of high-quality workshop here in Barcelona. Fortunately, I think the school here, as well as those nice organizers, really encourage students to take part in these kinds of seminars. Therefore, most time I can just walk in the conference room and pick a seat without registering in advance. Thank them so much for tolerating me -_-||

Oh one thing at last. Yesterday dear Ghazala also offered a seminar on her own research, which focused on gender difference in the labor market:

LABOUR/PUBLIC/DEVELOPMENT FACULTY LUNCH
Date: 23/03/2011
Speaker: Ghazala Azmat and Rosa Ferrer (UPF)
Title: Gender Inequality, Performance Pay and Young Professionals

Since it was her, it had no reason to be not good.Meanwhile, the active audiences were impressive as they had always been.

Moreover, today there was also another Chinese professor who talked about his economic history research,

ECONOMIC AND BUSINESS HISTORY SEMINAR SERIES
Date: 24/03/2011
Speaker: Debin Ma (LSE)
Title: Rock, Scissors, Paper: the Problem of Incentives and Information in Traditional Chinese State and the Origin of Great Divergence

I did not have enough time for this presentation, since the time was a little conflicted with Heckman's speech. However, I stayed there for half an hour and got the basic idea what was going on. It seemed that he was talking about the popular issue "Great Divergence (中文相当于李约瑟之谜)", and reminded me the time when I read Kenneth Pomeranz (彭慕兰)'s book last year:

  • Kenneth Pomeranz, The great divergence: China, Europe, and the making of the modern world economy (Princeton University Press, 2001).

When it comes to China, I cannot just stand there and say "I do not care", right? But the Great Divergence is not only a question on China, but also for the whole world. Institution or technology? All of us are curious which one drives the economy to grow on earth...

Physics > complex network > link prediction

It is a little funny that while I was trying to build the psychological-economic model in partner matching, two papers appeared in front of me, and they are all about link prediction:

  • Linyuan Lu y Tao Zhou, “Link Prediction in Complex Networks: A Survey,” 1010.0725 (Octubre 4, 2010), http://arxiv.org/abs/1010.0725.
  • D. Liben-Nowell y J. Kleinberg, “The link-prediction problem for social networks,” Journal of the American Society for Information Science and Technology 58, no. 7 (2007): 1019-1031.

Nowadays as an economics student, one important thing you can never overlook is math. That is why sometimes I feel that I am paying more attention to math instead of those economic thoughts. Yes we always stress economic intuitions. However, when it comes to the time problem, apparently math training acquires more time. More important, you may not able to build the entire economic intuition system only through lectures. Outside of the classroom is the place where you can really recognize the sense of how economics is related to people’s everyday life. Newspapers, street talks, and even TV shows may be more helpful.

Well… a little off topic. Today when I was listening to Thijs, suddenly a question not relative to any materials from that course came to my mind. What do we really care? Explanation, or prediction power? Of course they cannot be separated with each other. However, when it comes to prediction, it seems that people care more about the accuracy instead of why this or that method is working well. The best example may be the variety of methods of stock prices predicting. Once when I attended a statistics conference, there was someone who talked about his beautiful “triangle” prediction model of stock prices. For me, it does not make any sense. First, I’m not in the stock market, so I do not care; second, this kind of prediction methods are built above the cloud – there is no foundations which can even partly support the logic beyond the seemly accurate results. It may be unfair to attribute those meaningless methods to physicians – however, more of them studied physics before and they are somehow trying to transplant those solid physical theorems to economics, regardless the fact that economics is a branch of social science, which studies the (economic) functions of the society, the interactions among people and the fundamental reasons behind human behaviors. That is why whenever I see any model directly related to physics, I’m more careful about the economic implications behind the model: at least it should not be contra-intuitive.

Now myself is trying to borrow something from physics…. Obviously the first rule for me it to be as cautious as I can. In particular, I am trying to apply the link prediction idea to economic and social network analysis. Somewhat like the debate about reduced form and structural form in econometrics, I do not want to lose the micro foundations and economic intuitions behind the fancy model itself. It is easier if I just want to predict the result, but with consideration of internal and external validity, the model should be granted with more explanation power, which can conquer the difficulties originated from the complexity of the real society. Actually, I am fond of the name of this branch of science – complex network. How can we abstract simplicity from the complicated world? That may be the common question confronted by every single branch of science, no matter social science or hard/natural science. Without idea experiments in social science, how far can we go? It is really a difficult question to answer….

Anyway, hopefully I can find out a way to introduce link prediction ideas to my naive economic model. It is always interesting to work on the boundary of two different fileds, and you are tasting something new. However, the exams are coming so fast... need to focus on reviewing first.... Will update more later.

mother breastfeed and children health

This is an interesting topic: the behavior of mother breastfeed and children health. Yes, it is the topic of a seminar I audited yesterday.

As usual, here is the related information:

Why Do Mothers Breastfeed Girls less than Boys? Evidence and Implications for Child Health in India [click here to download PDF]
Seema Jayachandran (Stanford University)
UPF Labor, Public, and Development Seminar

Well, although I'm a kind of being addicted to theoretical works, this kind of empirical is still attractive.  She analyzed the development of children health with respect to mothor's breastfeed behaviors. As she said, it is like "quality instead of quantity", and mothers are paying more and more attention to ensure the health of their children.

Fortunately, one guy who lives nextdoor to me has been worked in the field of public health for the last year, and he has lived in India for several years, thus before the seminar he told me many things about the real situation in India. In this way, far beyond listening to the seminar itself, I have learned more about public health, or health economics which is really hot now.

The forever pain for an econometric or empirical analysis is the accuracy of data. She was using NFHS (National Family Health Survey) and DHS  (Demographic and Health Survey) data. However, my nextdoor told me that there were some problems in the questionnaire used, which will influence the reliability of the data, or the answer from those women. He doubted whether it would have potential influence on the regression analysis.

It was also interesting to hear that so many audience there had different kinds of interesting questions.  That is partically why I love economics seminars. But still the question for me is that I am still not able to follow all of the presentation. Apparently I need to store more knowledge.

By the way, I also listened to another seminar this week, it is

Misspecification in Models for Trends and Cycles
Andrew Harvey (University of Cambridge)
UPF Occasional Econometric Seminar

Well, it was a theoretical work, but it was theoretical economics, and I have no idea with that. Unfortunately, I went there early and sat down in the second row of the classroom. Therefore, I had to pretend that I was focusing on what he talked about, but in fact I could not understand anything except HP filter. Oh my poor econometrics, and my poor time series analysis. It also proves that I need to learn more to completely understand the top seminars. Carry on!

At last, I'd like to copy some conclusions from Semma's paper here.

  • the duration of breastfeeding negatively correlates with the mother's likelihood of a subsequent birth.
  • breastfeeding increases with birth order; if parents have a preference for sons, then boys are breastfed more than girls; children with older brothers are breastfed more; these gender e ects are smallest for high and low values of birth order
  • breastfeeding the current child helps prevent or delay a subsequent pregnancy, and a subsequent (perhaps unwanted) pregnancy often causes mothers to wean the current child.
  • If contraception crowds out breastfeeding, then policy makers might consider pairing contraceptive campaigns with promotion of breastfeeding or improvements in water quality. Conversely, if contraception enables a mother to breastfeed her children longer because she can space them further apart, then policies that expand access to contraception might have an added bene t of encouraging breastfeeding.

[Play Econometrics with R] Preview for Chapter 1-2 released!

I'm glad to announce that now the preview version for my brochure Play Economics with R has been released! This time I publish the first two chapters, which were just finished several days ago. At present it is written in Chinese, so there is only an English content for none-Chinese readers.

Content

Chapter 1 Get familiar with R

1.1 Data Import...................................... 6
1.2 Summary the data....................................... 7
1.2.1 Average value..................................... 7
1.2.2 Linear regression (ordinary least squares, OLS).................... 8
1.3 Plot a regreesion figure...................................... 9
1.4 Point prediction......................................... 10
1.5 Multiple linear regression..................................... 10
1.6 Save and edit the code.................................... 11
1.7 Search for help....................................... 11

Chapter 2 Start from cross-section data

2.1 Parameter test....................................... 12
2.1.1 t test..................................... 13
2.1.2 F test..................................... 13
2.2 Confidence Intervals....................................... 14
2.3 Dummy variables....................................... 14
2.3.1 grouped by the nature.................................. 14
2.3.2 grouped by the value................................. 14
2.3.3 interaction items................................... 15
2.3.4 specify the based group.................................. 17
2.4 Heteroscedasticity test...................................... 18
2.4.1 BP test (Breusch-Pagan Test)........................ 19
2.4.2 White test (White test for heteroskedasticity)................ 20
2.5 Robust standard deviation...................................... 20
2.6 Weighted least squares estimation (WLS).............................. 21
2.6.1 with disturbance form known ............................... 21
2.6.2 feasible generalized least squares (Feasible GLS, FGLS)................ 22
2.7 Generalized Linear Estimation (GLM)................................. 24
2.7.1 Maximum Likelihood Estimation (MLE).......... 25
2.7.2 Probit and Logit models.............................. 25
2.7.3 Tobit model................................... 26
2.7.4 Ordered Logit / Probit............. 27
2.8 Count Model............................... 28
2.8.1 Poisson Regression Model................... 28
2.8.2 test for dispersion.............................. 29
2.8.3 Negative binomial regression model............. 30
2.8.4 Zero-inflated Poisson model (ZIP)............ 30
2.9 Sample Selection Bias.................................... 32
2.9.1 Heckit model.................................. 32
2.10 Simultaneous Equations Model..................................... 33
2.10.1 two-stage least squares (2SLS) and instrument variable................... 33
2.10.2 Simultaneous Equations Model Estimation: Seemingly Unrelated Regression (SUR)... 34
2.11 Proxy Variables....... 35

Download

You may download it from GitHub: http://github.com/cloudly/Play-Econometrics-with-R/downloads (I'll keep updating this page)

Please tolerate the mistakes and typo, as well as some formatting problems I need to come over in the future.

News and Feedback

If you want to keep up with the latest news of this brochure, please send an email to: publication@cos.name or cloudlychen@gmail.com

Your feedback is welcome! Please also send to publication@cos.name or cloudlychen@gmail.com

What‘s Next?

I'm currently working on Chapter 4 Panel Data Analysis (so I skip Chapter 3 Time Series temporarily). Chapter 4 will include several interesting methods, like Fixed / Random Effect, Panel Data tests and GMM estimator, etc. Please tell me what you are looking for in my brochure and I'll add them if possible.