Recent Blog Posts

show advanced search
search criteria = ALL

Imagine Tom, your friend from back in school. When you parted with him on the day of graduation with promises of being in touch forever, which, at least then, implied individual initiatives beyond the horizons of convenience, little did you know that you’ll be later pulled back from your own verbal commitments by the very horizons that you planned to escape. The question is what happened?

For about a couple of years post school, you called each other regularly, and each remained updated with the tiniest detail of the other’s life. Things however began to change after that when you realized that your social circles were expanding … only too quickly for you to handle. Finally, the number of calls you made to each other, reduced to the level that only accommodated birthday greetings. What happened next?

Be the first person to recommend this.

The National Academies' Committee on Applied and Theoretical Statistics (CATS) released last week a summary of its popular April 2014 workshop, "Training Students to Extract Value from Big Data." You can view the video of the workshop here.

Co-chaired by John Lafferty of the University of Chicago and Raghu Ramakrishnan of the Microsoft Corporation and funded by the National Science Foundation, the workshop was convened to discuss how best to train students to use Big Data. As explained in the summary's introduction, the workshop explored four topics: 

  • The need for training in big data.
  • Curricula and coursework, including suggestions at different instructional levels and suggestions for a core curriculum.
  • Examples of successful courses and curricula.
  • Identification of the principles that should be delivered, including sharing of resources.
Be the first person to recommend this.

The University of Washington's Elizabeth Halloran and the University of Florida's Ira Longini are coauthors on two pieces relating to Ebola that were published yesterday.

In a letter to Science on which Halloran was lead author, Ebola: Mobility Data, the authors called for more data "to gain a more complete picture of mobility and infer patterns of disease spread." Noting "the current West African Ebola outbreak is taking place in a region where mobility has changed considerably in recent years," the authors state that additional "data can be used in dynamic transmission models to provide case projections, help focus resources and interventions, and assess the success of interventions." They conclude,

Such data should not necessarily lead to travel restrictions, such as flight route cancellations and border closures, which hamper relief efforts. Rather, the information should be used to create more valid models of transmission, which can then be used to plan and evaluate potential interventions. Better quantification of the impact of potential interventions will be critical in the coming weeks as the outbreak continues to grow.

Be the first person to recommend this.

[10/29/14: This blog was updated to include a University of Wisconsin BD2K grant with significant statistician involvement.]

NSF and NIH issued several press releases in the last couple months regarding high-profile solicitations to support administration initiatives. Statisticians were among the recipients. Here I profile a few of them, and one from the Moore Foundation, to highlight the work of statisticians. If I've missed any, please let me know.

The programs are:

  1. Big Data to Knowledge (BD2K), NIH
  2. Data Infrastructure Building Blocks (DIBBs), NSF
  3. Collaborative Research in Computational Neuroscience (CRCNS), NIH/NSF
  4. BRAIN Initiative, NSF & NIH   
  5. Gordon and Betty Moore Foundation
The October 9 NIH press release, "NIH invests almost $32 million to increase utility of biomedical research data," announced FY14 awardees of the NIH’s Big Data to Knowledge (BD2K) initiative, which, according to the press release, "is projected to have a total investment of nearly $656 million through 2020, pending available funds." JHU's Brian Caffo, Harvard's Rafael Irizarry, and University of Washington's Daniela Witten were awardees in the category of
Be the first person to recommend this.

The International Conference on Health Policy Statistics (ICHPS), sponsored by the ASA, will be held in Providence, RI from October 7 to 9, 2015. You and your coauthors/collaborators are encouraged to submit workshop and invited session proposals online until this Friday, October 20, 2014.
Be the first person to recommend this.

Be the first person to recommend this.

[10/30/14 update: Erosion of Federal Statistical Agencies Puts Sound Policy at Risk,Robert M. Groves and Kenneth Prewitt, Roll Call, 10/30/14.]

In a short September 17 markup, the Senate Health, Education, Labor, and Pension Committee approved H.R. 4366, the Strengthening Education through Research Act (SETRA), after minor modifications. As explained in the June 3 blog entry, House Passes Bill Diminishing Stature and Autonomy of National Center for Education Statistics; Senate Plans Unclear, the House-passed bill re-authorizing the Department of Education Institute of Education Sciences (IES) diminishes the stature and autonomy of the

Be the first person to recommend this.

ASA President-Elect David Morganstein sent letters to both House and Senate leaders in early September regarding H.R. 4012 and S. 2613both titled "Secret Science Reform Act of 2014"expressing concern and urging major revisions to the bills before further consideration. While generally applauding the bills' intent to make data underlying EPA rulemaking more available, Morganstein's letter (House version; Senate version) states,

Our concerns include those voiced by others (especially the American Association for the Advancement of Science) that the bill’s statements do not account for the complexities common to the scientific process on research that involves biological materials or physical specimens not easily accessible, combinations of public and private data, longitudinal data collected over many years that are difficult to reproduce, and data from one-time events that cannot be replicated...  We also agree with the point that it would be prudent to see the EPA’s data access policy—in accordance with the America COMPETES Reauthorization Act of 2010—expected by year’s end before further action on H.R. 4012.

Be the first person to recommend this.

At the current rate of development there will be about 10,000 R packages available in 2 years time! Click here to read more.
Be the first person to recommend this.

Employ your data skills for great causes!  


The Avalon Consulting Group is seeking a statistician to build and validate models and conduct data mining for direct marketing campaigns and program strategy.  Collaborate with our dynamic teams on award-winning campaigns for museums, theaters, and an exciting variety of progressive nonprofits and political campaigns.   

What You Will Do

1 person recommends this.

Original article here

Professor David Fox ponders on QA/QC issues for the R computing environment.
2 people recommend this.

Novavax, Inc. a clinical-stage biopharmaceutical company is seeking an Associate Director, Biostatistics to join their team in Gaithersburg, MD.


Responsibilities include:

• Assists clinical team in designing all NOVAVAX-sponsored clinical trials and prepares all statistical sections of clinical protocols using appropriate statistical methodology for the specific trial, including selection of study design, sample size, and analyses.

• Reviews database design, CRF's, and edit checks.

• Prepares statistical analysis plans.

Be the first person to recommend this.

The Office of Management and Budget (OMB) and the White House Office of Science and Technology (OSTP) released their joint budget priorities memo for FY16 on July 18. The joint memo is issued annually to guide federal agencies and departments in the preparation of their upcoming budget submissions. The document states, "Federal government funding for research and development (R&D) is essential to address societal needs in areas in which the private sector does not have sufficient economic incentive to make the required investments. Key among these is the fundamental, curiosity-driven inquiry that has been a hallmark of the American research enterprise and a powerful driver of unexpected, new technology." It also says, "Agencies should explain in their budget submissions how they are redirecting available resources from lower-priority areas to science and technology activities that address the priorities described below."

These memos are important for the statistical community to monitor because they provide opportunities to state how statisticians can contribute to national research priorities. In the past several months, three ASA groups released whitepapers relating to three national research priorities: the BRAIN Initiative, the Big Data R&D Initiative and the climate change research priority. The ASA whitepapers are described in this July
Be the first person to recommend this.

It is hard to believe it got here so fast, but the Joint Statistical Meetings are in full swing.  In my role as executive director, I'm making the rounds at the conference, updating members on some of the exciting activities of the ASA in its 175th anniversary year.  For those of you not able to join us in Boston, here is the summary I am sharing at various meetings:
Be the first person to recommend this.

Over the past 18 months, I've been writing about the growth in the number of statistics degrees being granted using data from the National Center for Education Statistics NCES. I've restricted my queries to five categories, which are represented by these five NCES CIP codes: 26.1102 – Biostatistics; 27.0501 - Statistics, General; 27.0502 - Mathematical Statistics and Probability; >27.0503 - Mathematics and Statistics; and 27.0599 - Statistics, Other. I generally use Statistics to be the last four categories, which NCES categorizes as 27.05 - Statistics. (See below for definitions and links.)

Because statistics is such a broad field and there are emerging areas like, for example, data science and analytics, it was suggested I use a blog entry to briefly discuss the CIP Codes and to consider statistics-related CIP codes. When one does look at the related degrees, one sees that some statistics-related degrees are also seeing strong growth.

Let me start with a chart of the four 27.05 categories, Biostatistics, and a related statistics degree: 27.9999 – Mathematics and Statistics – Other. For ease of comparison, the data below includes Bachelor's degrees categorized as 1st majors (and therefore misses another ~300 degrees.) One can see in the chart that most of Bachelor's degrees are categorized as 27.0501 - Statistics, General and that this category captures most of the growth in Bachelor's degrees. One will also notice a big jump in 27.0502 - Mathematical Statistics and Probability between 2011 and 2012, which is due to one university changing how it categorizes its Bachelor's degrees. It's also worth noting that the 27.9999 category has increased 50% since 2003.

Be the first person to recommend this.

The ASA Committee on Funded Research (CFR) will again be hosting a meeting at JSM for attendees to learn about funding opportunities directly from representatives of funding agencies. "Funding Opportunities for Statistics" is scheduled for Tuesday, August 5, 4-5:30 pm in room 157B of the Convention Center. This meeting has been very well attended in past years and I hope that you'll be part of it this year.

The confirmed speakers are

2 people recommend this.

The NIH Advisory Committee to the Director released a report in early June, BRAIN 2025: A Scientific Vision, that emphasizes the importance of statistics. Its executive summary lists seven scientific goals that are high priorities for achieving this vision including:

Identifying fundamental principles: Produce conceptual foundations for understanding the biological basis of mental processes through development of new theoretical and data analysis tools. Rigorous theory, modeling, and statistics are advancing our understanding of complex, nonlinear brain functions where human intuition fails. New kinds of data are accruing at increasing rates, mandating new methods of data analysis and interpretation. To enable progress in theory and data analysis, we must foster collaborations between experimentalists and scientists from statistics, physics, mathematics, engineering, and computer science.

This goal is then discussed in section 2.5, titled, "Theory, Modeling, and Statistics Will Be Essential to Understanding the Brain." That section listed and discussed several "areas that appear promising for the collaborative efforts of theorists and experimentalists:"
Be the first person to recommend this.

In keeping with the “celebrate our past, energize our future” theme of the 175th anniversary, we are pleased to announce today that “Statistics and Science: A report of the London workshop on the future of the statistical sciences” is posted on The World of Statistics website. The London workshop was the capstone meeting of the International Year of Statistics 2013.

Below is the blog I wrote for The World of Statistics website summarizing the report, but I urge you to read the full report. It is well worth the time!

Statistical science is as healthy as it ever has been, with robust growth in student enrollment, abundant new sources of data, challenging problems to solve and related opportunities to grasp over the next century, summarizes a just-released report on the future of the field.

Statistics and Science: A Report of the London Workshop on the Future of the Statistical Sciences ( is the product of a high-level meeting in London last November attended by 100 prominent statisticians from around the world. This invitation-only summit was the capstone event of the International Year of Statistics, a year-long celebration during 2013 that drew as participants more than 2,300 organizations from 128 countries.
Be the first person to recommend this.

We’ve written often in this 175th anniversary blog about the importance of communities within the ASA.  (See March 27, for example.)  Today we’ll focus on one specific type of community, a type many members are unfamiliar with, but one which serves an important purpose for those who are involved.

Interest groups are smaller ASA communities organized around an area of specific importance to portions of our diverse membership.  They are generally smaller than and more loosely organized than ASA sections, but they are every bit as important to their members.

There are four ASA interest groups.  The newest of these, Astrostatistics, was formed just a few months ago.  It was organized to meet a growing need in collaborative research efforts between statisticians and astronomers and to encourage astrostatistical research to flourish within the ASA.  The group hopes to draw more statisticians into astrostatistics, and, consequently, more astronomers into astrostatistics since those new statisticians will seek out astronomy collaborators.  They are connected with a Working Group on Astroinformatics and Astrostatistics of the American Astronomical Society.

Be the first person to recommend this.

[7/10/14 update: The ASA Submitted its comments July 9. Read the letter here.]

In a May 21 Federal Register notice, the Office of Management and Budget (OMB) solicits comments on a proposed new Statistical Policy Directive affirming "the fundamental responsibilities of Federal statistical agencies and recognized statistical units in the design, collection, processing, editing, compilation, analysis, release, and dissemination of statistical information." Comments are due July 21 and the ASA will be strongly supporting the proposed directive.

The proposed directive, which makes up two of the notice's five pages, delineates four responsibilities of federal statistical agencies and units to "provide a framework that supports Federal statistical policy and serves as a foundation for Federal statistical activities, promoting trust among statistical agencies, data providers, and data users." The four responsibilities are  

  1. Produce and disseminate relevant and timely information.
Be the first person to recommend this.