Decision Science News

DEADLINE JANUARY 10, 2013

Via Michael Birnbaum:

In this conference, investigators present topics that might be empirical or theoretical, involving questions that may be basic or applied, and studying theories that may be normative or descriptive. Topics deal with judgment and decision theory, basic and applied, either normative or descriptive, and are NOT limited to Bayes theorem or Bayesian statistics.

Daniel Cavagnaro and Michael Birnbaum will co-host the 51st meetings in Fullerton February 14-16, 2013. There will be a reception on the evening of February 14, with meetings on Friday and Saturday, Feb 15-16. Note that Monday, February 18 is “President’s Day”, which is a holiday in the USA and might permit an extra travel day for many American participants.

We hope you will accept the invitation to attend, which has more information on the conference.

Deadline to submit papers is January 10, 2013. On-Line Paper Submission and Registration for the 2013 Conference.

A Bit of History of these Meetings
The Bayesian Research Conference was hosted for 41 years by Ward Edwards, and for many years it was held each year in mid-February at the Sportsman’s Lodge in Studio City, California. These meetings have been very high in quality, extremely diverse in topic and approach, and under Ward’s leadership since 1961, they established a pattern of extremely fruitful, constructive, and congenial meetings. Starting with the 42nd meeting, the event has been held on the campus of California State University, Fullerton. picture of Ward Edwards and Jim Shanteau

The 44th and succeeding conferences are now the Edwards Bayesian Research Conference, honoring Ward Edwards, a founder of the field of behavioral decision research, who passed away in early 2005. In this picture, Ward and Jim Shanteau converse at an earlier meeting. As Ben Newall (2009) has written, “Ward Edwards is commonly regarded as the Father of behavioral decision making. In two papers, The Theory of Decision Making published in Psychological Bulletin in 1954 and Behavioral Decision Theory published in the Annual Review of Psychology in 1961 he founded and then subsequently gave a name to the field. Then in 1963 he introduced psychologists to Bayesian thinking with his paper Bayesian Statistical Inference for Psychological Research published in Psychological Review. These truly seminal papers have become part of the folklore of the field and their influence can hardly be overstated.” To those accomplishments, one might add his paper in the Psychological Review on Subjective probabilities inferred from decisions in which Edwards introduced the idea of probability weighting functions that depend on the configuration of consequences, an idea that is central to Prospect Theory, which was co-authored by his former student, Amos Tversky, and Daniel Kahneman.

The 2013 meeting will honor the memory of R. Duncan Luce, a leader in the field of behavioral decision making and a regular participant in this conference, who passed away in 2012.

ADDING DECISION-MAKING TRAINING INTO TRADITIONAL ACADEMIC COURSES

Chris Spetzler contributes this article, recently published in PLoS ONE, which finds that putting decision-making training into a a U.S. history course raised competence in both the material of the course, as well as in decision-making.

Title:

Improved Learning in U.S. History and Decision Competence with Decision-Focused Curriculum

Citation:

Jacobson D, Parker A, Spetzler C, Bruine de Bruin W, Hollenbeck K, et al. (2012) Improved Learning in U.S. History and Decision Competence with Decision-Focused Curriculum. PLoS ONE 7(9): e45775. doi:10.1371/journal.pone.0045775

URL:

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0045775

Abstract:

Decision making is rarely taught in high school, even though improved decision skills could benefit young people facing life-shaping decisions. While decision competence has been shown to correlate with better life outcomes, few interventions designed to improve decision skills have been evaluated with rigorous quantitative measures. A randomized study showed that integrating decision making into U.S. history instruction improved students’ history knowledge and decision-making competence, compared to traditional history instruction. Thus, integrating decision training enhanced academic performance and improved an important, general life skill associated with improved life outcomes.

Photo credit: http://www.flickr.com/photos/library_of_congress/3253742644/

MACHINE LEARNING, COMPUTATIONAL SOCIAL SCIENCE, ALGORITHMIC ECONOMICS, MARKET DESIGN AND MORE

Last week, we let you know about the Microsoft Research Postdoc in online experimental social science.

This week, we’re announcing four more postdocs! See below.

Microsoft Research NYC [ http://research.microsoft.com/newyork/ ] seeks outstanding applicants for 2-year postdoctoral researcher positions. We welcome applicants with a strong academic record in one of the following areas:

* Computational social science: http://research.microsoft.com/cssnyc
* Online experimental social science: http://research.microsoft.com/oess_nyc
* Algorithmic economics and market design: http://research.microsoft.com/algorithmic-economics/
* Machine learning: http://research.microsoft.com/mlnyc/

We will also consider applicants in other focus areas of the lab, including information retrieval, and behavioral & empirical economics. Additional information about these areas is included below. Please submit all application materials by January 11, 2013.

———-

COMPUTATIONAL SOCIAL SCIENCE
http://research.microsoft.com/cssnyc
With an increasing amount of data on every aspect of our daily activities — from what we buy, to where we travel, to who we know — we are able to measure human behavior with precision largely thought impossible just a decade ago. Lying at the intersection of computer science, statistics and the social sciences, the emerging field of computational social science uses large-scale demographic, behavioral and network data to address longstanding questions in sociology, economics, politics, and beyond. We seek postdoc applicants with a diverse set of skills, including experience with large-scale data, scalable statistical and machine learning methods, and knowledge of a substantive social science field, such as sociology, economics, psychology, political science, or marketing.

ONLINE EXPERIMENTAL SOCIAL SCIENCE
http://research.microsoft.com/oess_nyc
Online experimental social science involves using the web, including crowdsourcing platforms such as Amazon’s Mechanical Turk, to study human behavior in “virtual lab” environments. Among other topics, virtual labs have been used to study the relationship between financial incentives and performance, the honesty of online workers, advertising impact as a function of exposure time, the implicit cost of annoying ads, the testing of graphical user interfaces eliciting probabilistic information and also the relationship between network structure and social dynamics, related to social phenomena such as cooperation, learning, and collective problem solving. We seek postdoc applicants with a diverse mix of skills, including awareness of the theoretical and experimental social science literature, and experience with experimental design, as well as demonstrated statistical modeling and programming expertise. Must know and love R. Specific experience running experiments on Amazon’s Mechanical Turk or related crowdsourcing websites, as well as managing virtual participant pools is also desirable, as is evidence of UI design ability.

ALGORITHMIC ECONOMICS AND MARKET DESIGN
http://research.microsoft.com/algorithmic-economics/
Market design, the engineering arm of economics, benefits from an understanding of computation: complexity, algorithms, engineering practice, and data. Conversely, computer science in a networked world benefits from a solid foundation in economics: incentives and game theory. Scientists with hybrid expertise are crucial as social systems of all types move to electronic platforms, as people increasingly rely on programmatic trading aids, as market designers rely more on equilibrium simulations, and as optimization and machine learning algorithms become part of the inner loop of social and economic mechanisms. We seek applicants who embody a diverse mix of skills, including a background in computer science (e.g., artificial intelligence or theory) or related field, and knowledge of the theoretical and experimental economics literature. Experience building prototype systems, and a comfort level with modern programming paradigms (e.g., web programming and map-reduce) are also desirable.

MACHINE LEARNING
http://research.microsoft.com/mlnyc/
Machine learning is the discipline of designing efficient algorithms for making accurate predictions and optimal decisions in the face of uncertainty. It combines tools and techniques from computer science, signal processing, statistics and optimization. Microsoft offers a unique opportunity to work with extremely diverse data sources, both big and small, while also offering a very stimulating environment for cutting-edge theoretical research. We seek postdoc applicants who have demonstrated ability to do independent research, have a strong publication record at top research venues and thrive in a multidisciplinary environment.

SEEKING MATHEMATICALLY & COMPUTATIONALLY SKILLED APPLICANTS

Microsoft Research NYC seeks outstanding applicants with strong quantitative and programming skills for a postdoctoral researcher position in the area of online experimental social science.

Deadline for Full Consideration: January 11, 2013

Online experimental social science involves using the web, including crowdsourcing platforms such as Amazon’s Mechanical Turk, to study human behavior in “virtual lab” environments. Among other topics, virtual labs have been used to study the relationship between financial incentives and performance, the honesty of online workers, advertising impact as a function of exposure time, the implicit cost of “bad ads”, the testing of graphical user interfaces eliciting probabilistic information and also the relationship between network structure and social dynamics, related to social phenomena such as cooperation, learning, and collective problem solving. Eligible applicants must hold a Ph.D. in Computer Science, Experimental Economics, Experimental Psychology, Statistics, Mathematical Sociology or a related field. The ideal applicant will possess a diverse mix of skills, including awareness of the theoretical and experimental social science literature, and experience with experimental design, as well as demonstrated statistical modeling and programming expertise. Programming knowledge should include server-side and browser-side languages, interaction with databases and third party APIs and facility with the R language for statistical computing. Specific experience running experiments on Amazon’s Mechanical Turk or related crowdsourcing websites, as well as managing virtual participant pools is also desirable, as is evidence of UI design ability. Postdoc researcher positions at Microsoft Research provide emerging scholars (Ph.D.s received in 2012 or to be conferred by July 2013) an opportunity to develop their research career and to interact with some of the top minds in the research community. The position also offers the potential to have research realized in products and services that will be used worldwide. Postdoc researchers are invited to define their own research agenda and demonstrate their ability to drive forward an effective program of research.

Postdoc researchers receive a competitive salary and benefits package, and are eligible for relocation expenses. Postdoc researchers are hired for a two-year term appointment following the academic calendar, starting in July 2013. Applicants must have completed the requirements for a Ph.D., including submission of their dissertation, prior to joining Microsoft Research. We do accept applicants with tenure-track job offers from other institutions so long as they are able to negotiate deferring their start date to accept our position.

About MSR-NYC
Microsoft Research provides a vibrant multidisciplinary research environment with an open publications policy and close links to top academic institutions around the world. Microsoft Research New York City is the most recent MSR lab, comprising 16 full-time researchers and postdocs, working on theoretical and applied aspects of machine learning and information retrieval, computational and online experimental social science, and algorithmic and experimental economics. The lab is highly collaborative and interdisciplinary, and its members also maintain active links both with the local academic and tech communities.

For more information about the lab, visit:

http://research.microsoft.com/en-us/labs/newyork/default.aspx

To apply for a postdoc position at MSR-NYC:

1. Submit an online application at:

https://research.microsoft.com/apps/tools/jobs/fulltime.aspx

* Indicate that your research area of interest is “Online Experimental Social Science” and that your location preference is “New York.” Include the name of a Microsoft Research contact if you have one.

* In addition to the CV and names of three referees (including your dissertation advisor) that the online application will require you to include, upload the following 3 attachments with your online application: a) two conference or journal articles, book chapters, or equivalent writing samples (uploaded as 2 separate attachments); b) an academic research statement (approximately 3-4 pages) that outlines your research achievements and agenda.

2. After you submit your application, send an email to msrrt@microsoft.com (copy the Microsoft Research contacts you identified in step 1, if any) alerting us that you have uploaded your application. If an applicant meets the requirements above, a request for letters will be sent to your list of referees on your behalf. All letters of recommendation must be received by the deadline for full consideration of the application. Please make sure to check back with your referees or us if you have any questions about the status of your requested letters of recommendation. For more information, see:

http://research.microsoft.com/en-us/jobs/fulltime/postdoc.aspx

NOT SURE WHY DAYLIGHT SAVINGS TIME IS WHEN IT IS

Last week we posted a nice theory about daylight savings time, in particular, that its dates were chosen to reduce variance in the time of sunrise. It looked plausible from the graph.

We were talking to our Microsoft Research colleague Jake Hofman who suggested “why don’t you just find the optimal dates to change the clock by one hour?” So we did. We got the times of sunrise for New York City from here, threw them into R, and optimized.

The result was surprising. The dates of daylight savings time do not come close to minimizing variance in sunrise. If they did, in 2012, DSL would have started on March 25th and ended on September 28th. In actuality, it started on March 11th and ended on November 4th. For NYC, daylight savings time starts too early and ends too late to minimize variance in sunrise. In the heatmap above, the higher the variance, the bluer the squares. The variance minimizing dates are shown in black, and the actual ones in red. The same color coding is used in the plot below, which also shows how the hours would shift if the variance minimizing dates were chosen (see last week’s post for how they actually change).

So what then is the logic behind DSL? We’re not quite sure. There are some leads in this article. We also learned that the US lengthened DSL in 2007 as it believes it that DSL saves energy, but it is not clear that it does.

If you want to play with this, the data are here: Sunrise and Sunset data for New York City in 2012. The source of the data is here.


library(ggplot2)
#data from http://aa.usno.navy.mil/data/docs/RS_OneYear.php
setwd("C:/Dropbox/Projects/Sunrise/")
df=read.table("nyc_sunrise.txt",colClasses="character")
temp_times=unlist(df[,c(paste("V",seq(2,24,2),sep=""),
paste("V",seq(3,25,2),sep=""))])
l=length(temp_times)/2
df=data.frame(day=1:31,stime=temp_times,
sun=c(rep("rise",l),rep("set",l)))
rm(temp_times,l)
hour=as.numeric(substr(df$stime,1,2))
minute=as.numeric(substr(df$stime,3,4))/60
df$time=hour+minute
df=subset(df,!is.na(df$time))
df$day_of_year=1:(nrow(df)/2)
p=ggplot(data=subset(df,sun=="rise"),aes(x=day_of_year,y=time))
p=p+geom_line()
p=p+geom_line(data=subset(df,sun=="set"),aes(x=day_of_year,y=time))
p
zerovec=function(i,j){
c(rep(0,i-1),
rep(1,j-i+1),
rep(0,len-j))}
currvec=df[df$sun=="rise","time"]
len=length(currvec)
get_var=function(i,j) {
var(currvec +
zerovec(i,j))}
vget_var=Vectorize(get_var)
result=expand.grid(spring_forward=45:125,fall_back=232:312)
result=subset(result,spring_forward<fall_back)
result$var=with(result,vget_var(spring_forward,fall_back))
resout=result[which.min(result$var),]
resout
#Heatmap
p =ggplot(data=result, aes(spring_forward, fall_back)) +
geom_tile(aes(fill = var), colour = "white") +
scale_fill_gradient(low = "white", high = "steelblue")
p=p+geom_vline(xintercept=as.numeric(resout[1]))+
geom_hline(yintercept=as.numeric(resout[2]))
p=p+geom_vline(xintercept=71,color="red")+
geom_hline(yintercept=309,color="red")
p=p+ylab("Fall Back Day of Year\n")+theme_bw()
p=p+xlab("\nLeap Forward Day of Year")+opts(legend.position="none")
p
ggsave("heatmap.pdf",p,width=6)
#For 2012, optimal spring forward day (85)is 03/25/2012
#For 2012, optimal fall back day (272) is 09/28/2012
#Actual DSL start was 3/11/2012 (71)
#Actual DSL end was 11/4/2012 (309)
p=ggplot(data=subset(df,sun=="rise"),
aes(x=day_of_year,y=time+zerovec(85,272)))
p=p+geom_line()
p=p+geom_line(data=subset(df,sun=="set"),
aes(x=day_of_year,y=time+zerovec(85,272)))
p=p+geom_vline(xintercept=85,lwd=2)+
geom_vline(xintercept=272,lwd=2)
p=p+geom_vline(xintercept=71,color="red")+
geom_vline(xintercept=309,color="red")
p=p+ylab("Hour")+theme_bw()
p=p+xlab("\nDay Of Year")+opts(legend.position="none")
p
ggsave("timeshift.pdf",p,width=6)


Figures created with Hadley Wickham’s ggplot2

ADJUSTMENT TO REDUCE VARIANCE IN THE TIME OF SUNRISE

How did they decide when and by how much to make the “daylight savings time” adjustment? There’s a nice visualization at http://visual.ly/daylight-saving-time-explained which suggests that daylight savings time was chosen in a way to keep the time of sunrise relatively constant throughout the year. Check out the chart above: the time of sunrise deviates from the “average sunrise” time less than the time of sunset deviates from the “average sunset” time. Compelling.

The creator of the visualization, who seems to go by the handle germanium, writes

I wanted to see the effect of daylight saving time change on sunrise and sunset times. The data was taken from http://www.timeanddate.com and is for Chicago. The figure shows that daylight saving time change (marked by the DLS lines) keeps the sunrise time pretty much constant throughout the whole year, while making the sunset time change a lot. The spread of sunrise times for the whole year as measured by the standard deviation is 42 minutes, while for sunset times is 1:30 hours.

DON’T STORE SMALL CHANGE, PUT IT TO USE

A friend recently asked us what to do with his large and unintentional collection of foreign coins left over from many international trips. He was surprised to learn that currency exchanges won’t take them. Since hanging onto coins between foreign trips is annoying (*), we recommend the following three practices for putting your coins to purposeful use.

1) Give them to charity. There are collection boxes in most international airports. Once you know about them, you’ll see them everywhere. We took the pictures in this post on just one international trip.

2) Use them plus a credit card to buy something at an airport shop. A few bucks worth of foreign currency won’t buy anything interesting at a duty-free shop, however, the same shops will happily accept all your small change plus a credit card to make up the difference on a larger item. We often buy things we are going to use anyway, you know, like batteries and gin.

3) If you have former European currencies that are no longer circulating (Belgian franc, Deutsche Mark, Estonian kroon, Irish pound, Luxembourg franc, Maltese lire, Dutch Guilder, Austrian Schilling, Portuguese escudos, Slovak koruna, Slovenian tolar, Spanish pesetas und Cypriot pound), you’re not entirely out of luck. You can send them to euromoney24.com and ask that they be donated to charity or transferred to your bank account.

(*) We do hold on to bills, however, for countries we expect to revisit within a couple years

HIDDEN HINTS HELP DRIVERS

In response to last week’s surprisingly popular “how to tell which side the gas cap is on”, we received a number of emails, one by our friend and co-author Preston McAfee, who shares a tip [from this site] that also helps driving by letting you know in advance on which side your exit will be. The picture says it all.

Photo credit: http://shialabeowulf.tumblr.com/post/33670447154/99-life-hacks-to-make-your-life-easier

DRIVE THAT RENTAL CAR TO THE CORRECT SIDE OF THE PUMP

Somewhere along the road of life, we at Decision Science News learned that the little triangle next to the fuel gauge points to the gas-cap side of the car.

We’re not promising this works on every car, but it’s worked on every car we’ve rented.

Guess you’ll need to find some other occasion to practice your backing-up-while-turning skills.

A TOOL TO REPRESENT PROBABILITIES IN AN EASY-TO-UNDERSTAND WAY

We’ve written before about using information grids when communicating risks to the general public. We like them. Turns out they are also called pictographs and, as we learned from an email from Brian Zikmund-Fisher, icon arrays.

As Brian puts it

Iconarray.com is a free online tool that enables anyone to create icon array risk graphics (sometimes called pictographs), download the images, or even embed the graphics in web pages just the way you can embed YouTube videos.

To learn more, see my blog post at Risksense.org or just visit the site directly at iconarray.com.

The current demonstration version is sponsored by the University of Michigan Risk Science Center and the UM Center for Bioethics and Social Sciences in Medicine.

Here’s a glimpse at the user interface: