February

February

  
Parameter
The Official Newsletter of the Chicago Chapter of the American Statistical Association
February 2015
In This Issue
Although winter began with a relatively innocuous start, these past few weeks have assured us that, yes indeed, it is winter here in the Midwest, with piles of snow to spare! Provided you're not still digging your vehicle out of a snowbank, we have some great events coming up in the next few months.
Registration is now open for our March conference on Prediction & Forecasting! Topics are to include economics, marketing, climate research, genetics, and sports, with speakers including Scott Brave (who is also our February luncheon speaker), Kenneth Kahn, Marker Berliner, Haky Im, and Allen Sanderson.
 
Last year our conference was our most popular event and sold out. Register now - don't miss out on the event of the season!
Continue reading for more about our February luncheon with Scott BraveMixed frequency state-space models and their applications: A review of Chicago Fed indexes, which will discuss general issues faced in modeling common mixed frequency economic and financial data series in a linear state-space framework.
We've also included articles about the statistical side of the recent Super Bowl, the accelerating growth of Statistics undergrad degrees, LinkedIn's hottest skills that got people hired, and information about how to advertise your job opening, academic program, or consulting services with the CCASA.



Coming Next Tuesday
February Luncheon

Mixed Frequency State-Space Models 
and Their Applications:
A Review of Chicago Fed Indexes
Scott Brave

February 17th 2015
12pm-1:30pm
East Bank Club
500 N. Kingsbury
Chicago, IL 60610


Abstract
To support its efforts to conduct US monetary policy, the Federal Reserve collects a large amount of data on economic and financial activity. Research departments within the Fed system disseminate much of this data to the public on a regular basis along with current research describing its applications. For instance, the Federal Reserve Bank of Chicago releases four indexes (the CFNAI, NFCI, MEI, and CFMMI) with applications to forecasting economic and financial conditions.
The construction of each of these four indexes in various ways makes use of recent advances in mixed frequency state-space modeling and estimation. Mr. Brave will discuss the general issues faced in modeling common mixed frequency economic and financial data series in a linear state-space framework. Applications of the Chicago Fed indexes to be considered include"nowcasting" gross domestic product (GDP) growth and the real-time monitoring of financial stability.
Bio
Scott Brave is a senior business economist in the economic research department of the Federal Reserve Bank of Chicago. His chief responsibilities include the releases for the Chicago Fed's National Activity and Financial Conditions Indexes, Midwest Economy and Manufacturing Indexes, and the Seventh Federal Reserve District's contribution to the Beige Book. Brave received a B.A. in economics with honors from the University of Chicago and an M.B.A. with concentrations in economics, statistics and finance from the University of Chicago Booth School of Business.
Registration Now Open
Prediction & Forecasting Conference
We are hosting our annual conference on March 6. The theme of this year is 'Prediction and Forecasting', a hot topic in our age of data deluge. Speakers will showcase how to use prediction and forecasting methods in the world of economics, marketing, climate research, genetics, and sports
Speakers are Scott Brave (Federal Reserve Bank of Chicago), Kenneth Kahn (Virginia Commonwealth University), Marker Berliner (Ohio State University),Haky Im (University of Chicago), and Allen Sanderson(University of Chicago). We hope that many of you will take advantage of this annual event.
Mark Your Calendars
March Luncheon
Is There Always a First Time?
Albert Madansky

March 17th 2015
12pm-1:30pm
East Bank Club
500 N. Kingsbury
Chicago, IL 60610


Abstract 

Chapter 9 of Sharon Bertsch McGrayne's 2011 book,The Theory That Would Not Die, describes my RAND Corporation classified research in 1958 "On the Risk of an Accidental or Unauthorized Nuclear Detonation." Some of my analysis was Bayesian in nature, and so Sharon McGrayne learned of it (this work was not declassified until July 2000) and decided to include it in her book on Bayes' Rule, deftly titling her chapter "There's Always a First Time." In this talk I will describe the statistical methods I used in determining the probability of a hitherto unexperienced event occurring in the future.

 

Bio 

Albert Madansky is the H.G.B. Alexander Professor of Business Administration Emeritus in the University of Chicago Booth School of Business. He has been a faculty member at Chicago Booth since 1974, and in addition to teaching a variety of courses in statistics and econometrics, he has taught courses in business economics, information science, marketing, business policy, as well as a course entitled "Great Books in Business."

He also served as director of the Center for Management of Public and Nonprofit Enterprise, director of the Center for International Business and Education Research, associate dean for PhD studies, and deputy dean for Faculty. In addition, he was editor-in-chief of its Journal of Business.

 

 

Registration Now Open
Spring Workshop 2015
Modern Regression Strategies
Frank E. Harrell Jr.
Vanderbilt University

April 24th 2015
8:30am - 4:30pm
The Feinberg School of Medicine
Northwestern University
680 North Lake Shore Drive - Ste 1400
Chicago, IL 60611


Abstract
All standard regression models have assumptions that must be verified for the model to have power to test hypotheses and for it to be able to predict accurately.Of the principal assumptions (linearity, additivity, distributional), this course will emphasize methods for assessing and satisfying the first two. 

Practical but powerful tools are presented for validating model assumptions and presenting model results. This course provides methods for estimating the shape of the relationship between predictors and response using the widely applicable method of augmenting the design matrix using restricted cubic splines.

Who Should Attend: 
This program will benefit all analytics professionals, especially those who want to get a good foundation in regression modeling in practice. Note: wireless access will be available at the venue, participants are encouraged to bring laptops or tablets.

Bio:
Dr. Frank Harrell is Professor of Biostatistics and Chair of the Department of Biostatistics of Vanderbilt University School of Medicine, Nashville, TN. He received his Ph.D. in Biostatistics from the University of North Carolina, Chapel Hill in 1979. Dr. Harrell's interests include statistical modeling and model validation, statistical computing and graphics, reproducible research, survival analysis, clinical trials, health services and outcomes research, medical diagnostic and prognostic models, bootstrapping, missing data, and Bayesian modeling.




An Important Data Lesson
from an Inconsequential
Football Scandal
Those of you that watched the Super Bowl this year may have heard of the "scandal" that erupted in the weeks before it involving the New England Patriots and deflated footballs, which the media promptly christened "DeflateGate", and then proceeded to report it to death.

Opinions on the media cycle aside, one of the unfortunate circumstances to rise out of the scandal were some truly tortured and wildly inaccurate uses of statistics, including, as this article from the Harvard Business Review astutely points out, some causal leaps inferred from correlations in the data. The article drives home an important lesson that our media seems to frequently ignore: correlation does not equal causation.



Ethical Guidelines for
Statistical Practice
On that note, "DeflateGate" serves as a great reminder that the American Statistical Association has Ethical Guidelines for Statistical Practice.

Be sure that you, your students or clients, and others are familiar with the professional responsibilities in our field - together we can (hopefully) put a halt to similar statistical disasters!



Statistics is the Fastest-Growing
Undergrad STEM Degree
The ASA analyzed data compiled by the National Center for Education Statistics (NCES) and found that Statistics is the fastest-growing undergrad STEM degree, even ahead of Computer Science. Overall, the number of statistics bachelor's degrees has grown from 526 in 2003 to 1,678 in 2013.

Surprised? Probably not. The Big Data/data science hype is certainly having an effect on education programs, and no doubt influencing students to choose Statistics due to the hot job market. It probably doesn't hurt that "statistical analysis and data mining" were ranked #1 on LinkedIn's list of the 25 hottest skills.



Advertising with the CCASA
Would you like to advertise a job posting, academic program, or consulting services in the Chicago Chapter of the American Statistical Association's monthly newsletter and on our website? 
Our Parameter goes out to thousands of quantitative professionals, students, and academics in the Chicagoland area and is a great option for targeted advertising to the statistical community. Advertisements remain valid for one month.

Please visit our information page for more details on payment and ad specifications.