Tag Archives: economics

recent research trend of NBER working papers, since 2013

Recent days I got curious that what topics attract most of economists' attentions. NBER working paper series contain some relatively new research fruits so I used it as the raw input.

It is not hard to extract key words from these papers' titles. After that,  I made a further step that matched all single keys to academic keywords on Microsoft Academic.

NBER working paper trend

From a glance, I manage to identify some hot keys:

Price. Health. Social. Policy/Public. Risk/asset/liquidity. Growth. Insurance. Education/School.

For reproducible purpose, my code is here.

 

grab_url <- c("http://www.nber.org/new_archive/mar14.html",
              "http://www.nber.org/new_archive/dec13.html",
              "http://www.nber.org/new_archive/sep13.html",
              "http://www.nber.org/new_archive/jun13.html",
              "http://www.nber.org/new_archive/mar13.html")

library(RCurl)
require(XML)

grab_paper <- function (grab) {
  webpage <- getURLContent(grab)
  web_content <- htmlParse(webpage,asText = TRUE)
  paper_title <- sapply(getNodeSet(web_content, path="//li/a[1]"),xmlValue)
  author <- sapply(getNodeSet(web_content, path="//li/text()[1]") ,xmlValue)
  paper_author <- data.frame(paper_title = paper_title, author = author)
  return(paper_author)
}

library(plyr)
paper_all <- ldply(grab_url,grab_paper)

titles <- strsplit(as.character(paper_all$paper_title),split="[[:space:]|[:punct:]]")
titles <- unlist(titles)

library(tm)
library(SnowballC)
titles_short <- wordStem(titles)
Freq2 <- data.frame(table(titles_short))
Freq2 <- arrange(Freq2, desc(Freq))
Freq2 <- Freq2[nchar(as.character(Freq2$titles_short))>3,]
Freq2 <- subset(Freq2, !titles_short %in% stopwords("SMART"))
Freq2$word <- reorder(Freq2$titles_short,X = nrow(Freq2) - 1:nrow(Freq2))
Freq2$common <- Freq2$word %in% c("Evidenc","Effect","Econom","Impact","Experiment","Model","Measur","Rate","Economi",
                                  "High","Data","Long","Chang","Great","Estimat","Outcom","Program","Analysi","Busi"
                                  ,"Learn","More","What")
library(ggplot2)
ggplot(Freq2[1:100,])+geom_bar(aes(x=word,y=Freq,fill = common,alpha=!common))+coord_flip()

### get some keywords from Bing academic
start_id_Set = (0:5)*100+1
require(RCurl)
require(XML)
# start_id =1
# 

get_keywords_table <- function (start_id) {
  end_id = start_id+100-1
  keyword_url <- paste0("http://academic.research.microsoft.com/RankList?entitytype=8&topDomainID=7&subDomainID=0&last=0&start=",start_id,"&end=",end_id)
  keyword_page <- getURLContent(keyword_url)
  keyword_page <- htmlParse(keyword_page,asText = TRUE)
  keyword_table <- getNodeSet(keyword_page, path="id('ctl00_MainContent_divRankList')//table")
  table_df <- readHTMLTable(keyword_table[[1]])
  names(table_df) <- c("rowid","Keywords"   ,  "Publications" ,"Citations")
  return (table_df)
}

require(plyr)
keywords_set <- ldply(start_id_Set,get_keywords_table)

save(keywords_set, file="keywords_set.rdata")

 

[Publication] Perhaps...the first publication under BGSE in Chinese

At the end of last month, I was notified that one of my papers co-authored with my previous advisor, Yue Qiao, was accepted. Today I downloaded an electronic version of it and finally confirmed.

That paper is published with the title The Mechanism Evolution and Information Transmission in Online Markets (我国网上交易的机制演化与信息传导) , at the Journal of Zhongnan University of Economics and Law, 2012, 192 (03). At the moment when we submitted the paper, I was studying at the Barcelona Graduate School of Economics, so interestingly the institution behind my name is BGSE in the published version.

I should say it is an old paper - and I have waited for three years to get it published. It is not a short time period - as I have graduated and have already changed a job. On the other hand, the paper is about mechanisms in online C2C market (Taobao.com as the context), and now I am working at eBay and am committed to C2C behavior analysis. Sounds like a regression, right?

Perhaps it is the version reason why I chose eBay immediately right after I got their offer. Theories need to be examined in the real market environment, and eBay is exactly the right place to do this! I have no reason not to be excited about working here. As the theoretical research is accepted, I should spend more time on the empirical part now. Hopefully, these years of knowledge gaining and work experience will enhance my analytical thinking ability and get a fruitful result afterwards.

I'm going on the way, and there is no word called "give up" in my dictionary.

The print version of this paper in Chinese can be download here: Mechanism_evolution_C2C.pdf

note on beliefs and expectation

Yesterday and today I spend several hours on the conference held by Crei,

Information, Beliefs and Expectations in Macroeconomics

20 & 21 May 2011

Organized by Kristoffer Nimark (CREI & UPF) and Bartosz Maćkowiak (European Central Bank).

To be honest, I have waited for the conference for a while, as I was writing things on beliefs and expectations. However, this conference is a little bit out of my expectation - so again, my expectations deviates from the truth. My poor background in macro is definitely not enough to support me for understanding the presentations, and among all of them, I picked a few to stay and thus make myself confused. They were:

Exogenous Information, Endogenous Information, and Optimal Monetary Policy
Luigi Paciello | Einaudi Institute for Economics and Finance
*Mirko Wiederholt | Northwestern University
Discussant: Jordi Galí | CREI & Universitat Pompeu Fabra

Learning about Consumption Dynamics
Michael Johannes | Columbia U., Columbia Business School
Lars Lochstoer | Columbia U., Columbia Business School
*Yiqun Mou | Columbia U., Columbia Business School
Discussant: Francisco Barillas | Emory U., Goizueta Business S.

Public’s Inflation Expectations and Monetary Policy
Leonardo Melosi | London Business School
Discussant: Francesco Bianchi | Duke University

Just as I had expected, it was too easy for me to get lost in these presentations. In a few minutes I got the feeling that I don't know what's going on so ever. well...Except for the fantastic graphs I have seen, I haven't gained much intuition from them.

A noteworthy point is in Mou's presentation, he empirically showed the convergence to rational expectation. I should admit I don't really understand the techniques he utilized, and so are the debates afterwards. After his presentation, I talked to him with the learning process a little bit, but haven't benefited a lot... there is still a long way to go.

Also, I found another interesting book to read,

1587, a year of no significance : the Ming dynasty in decline / Ray Huang

Very nice book on history. I really treasure this period of time that I have got enough space to read and think across different subjects. As planned before, I also found Keynes' book,

A Treatise on probability

Things are so beautiful!

[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.