http://forums.flowingdata.com/ Strength in Numbers en Sat, 11 Feb 2012 20:07:50 +0000 http://forums.flowingdata.com/topic/senior-biostatistician-clinical-research-kai-research-maryland#post-2488 Thu, 12 Jan 2012 14:07:34 +0000 bpiper 2488@http://forums.flowingdata.com/ <p>Job ID: KAI-PS-03-11<br /> <a href="https://jobs-altarum.icims.com/jobs/2027/job" rel="nofollow">https://jobs-altarum.icims.com/jobs/2027/job</a></p> <p>Overview:<br /> KAI Research, an Altarum company, is seeking a Biostatistician to work in our Clinical Research Strategies and Systems Practice Area in Rockville, MD. The Biostatistician will be responsible for planning, analyzing, reviewing, interpreting and summarizing the statistical component of research projects in compliance with regulatory guidelines and requirements.</p> <p>Responsibilities:<br /> - Provide sound strategic, scientific and statistical input to support our clients' research projects.<br /> - Take lead responsibility for the scientific integrity of the clinical studies and the accuracy of the reported results.<br /> - Review project protocols and case report forms to ensure protocol objectives are met and standards are maintained and generation of treatment assignments in randomized clinical trials is accurate.<br /> - Write the protocol statistical section, incorporating appropriate statistical methods for randomization, study design, endpoints and analysis.<br /> - Lead the study teams’ development of analysis plans and statistical programs to perform analyses and display study data.<br /> - Author the results sections of the clinical study report and/or scientific publication.<br /> - Represent KAI at meetings of clinical investigators, Data and Safety Monitoring Committees, and FDA or other regulatory meetings.<br /> - Engage in discussions and research, both inside and outside of KAI, to improve statistical or clinical trials methodology.<br /> - Participate in writing submissions to regulatory agencies, scientific papers and grant//contract proposals.<br /> - Review and analyze clinical research projects with respect to scientific scope, study procedures and human subject issues.<br /> - Assist in the development of clinical trials guidelines. </p> <p>Qualifications:<br /> - PhD and at least 3-5 years experience with clinical trials.<br /> - Good working knowledge of SAS and preferably S plus or R for statistical analyses.<br /> - Ability to move across studies and provide other study support<br /> - Experience dealing with clients and medical/scientific community.<br /> - Excellent written and verbal communication skills.<br /> - Ability to work both independently and in a matrix team environment.<br /> Strong leadership and interpersonal skills are required. </p> <p>ABOUT ALTARUM: </p> <p>Altarum's dynamic work environment emphasizes integrity, personal commitment, and teamwork. In addition to the meaningful and challenging work, Altarum offers an outstanding benefits program that includes paid time off, tuition reimbursement, 401k (5% employer contribution), a flexible work environment, and much more. We invite you to explore exciting and rewarding career opportunities with us. </p> <p>Altarum is an Equal Employment Opportunity/Affirmative Action Employer M/F/D/V. </p> http://forums.flowingdata.com/topic/real-time-statistical-modellersoftware-developer-london#post-2288 Mon, 01 Aug 2011 11:00:01 +0000 SamBrown 2288@http://forums.flowingdata.com/ <p>Statistical Modeller or Software Developer required in central London to work on challenging real time predictive modelling and Data Mining projects. This will be a fast paced, challenging role where you will take a lead role in statistical modelling, machine learning and data mining research, developing insightful analyses of multiple, disparate, massive, real-time data sets. You will be working closely with software engineers to integrate your algorithms into the backend systems and also with data visualisation specialists to help present results in creative, intuitive ways. You will be working with statistics and machine learning packages such as R, Matlab, Octave and Weka, along with implementing algorithms in C or C++. If you are looking for a challenging role involving statistical modelling, machine learning and software development working with massive data sets then please apply now for more details.</p> <p>The ideal candidate would be educated to Masters or PhD level in a relevant discipline such as statistics, mathematics, engineering, physics or possibly computer science and would have proven commercial (ie not just academic) experience in statistical modelling and data mining. You should have experience working with very large data sets and should have experience gathering data from disparate sources. You should have experience in at least one mainstream programming language such as C, C++ or Java. You must also have experience working with statistics, data mining or machine learning packages such as R, Matlab, Octave or Weka. Experience with Linux, SQL databases, data visualisation techniques, natural language processing or text categorisation would also be desirable.</p> <p>This is a great opportunity for an experienced statistical modeller or data mining expert to join a highly successful, growing company working on large, challenging projects.</p> <p>To apply, please contact samdotbrownatengagepsgdotcodotuk or visit <a href="http://www.engagepsg.co.uk/statistical-modeller-software-developer--data-mining-c-148-p-2928" rel="nofollow">http://www.engagepsg.co.uk/statistical-modeller-software-developer--data-mining-c-148-p-2928</a>. </p> http://forums.flowingdata.com/topic/suggest-the-software#post-1974 Tue, 02 Nov 2010 13:14:53 +0000 nathany 1974@http://forums.flowingdata.com/ <p>If you want to avoid programming and put it online, Tableau Public could be a good place to start. This list of resources also might be helpful:</p> <p><a href="http://flowingdata.com/2008/10/20/40-essential-tools-and-resources-to-visualize-data/" rel="nofollow">http://flowingdata.com/2008/10/20/40-essential-tools-and-resources-to-visualize-data/</a> </p> http://forums.flowingdata.com/topic/suggest-the-software#post-1973 Tue, 02 Nov 2010 10:41:27 +0000 Przemo 1973@http://forums.flowingdata.com/ <p>Hello Everyone,<br /> I am not experienced person in creating web-applications,<br /> however I have to present data analysis results (in the most dynamic way)<br /> on the website. The data so far is in postgressql, and problem is which software and<br /> languages for webapps are the most suitable to generate dynamic reports.<br /> The analysis includes mostly basic descriptive statistics (mean,median, quartiles, sd, etc.) and plots (mostly line charts). It has hundreds strata hence I'd like to have it dynamic, when user can select one variable and see the results of analysis.<br /> I am sorry in advance, if my question is stupid, but I am not experienced programmer.<br /> Is there any framework available?<br /> Thank you!<br /> p. </p> http://forums.flowingdata.com/topic/tracking-music-play-statistics-via-itunes#post-1956 Tue, 19 Oct 2010 15:51:42 +0000 JuliaJ 1956@http://forums.flowingdata.com/ <p>Have you seen this? <a href="http://www.leebyron.com/what/lastfm/" rel="nofollow">http://www.leebyron.com/what/lastfm/</a></p> <p>I absolutely love the idea of a stream graph and it's intriguing to see how musicians can fade and reappear across time. Just a thought.</p> <p>Julia </p> http://forums.flowingdata.com/topic/search-engine-marketing-analyststatistician-london#post-1953 Mon, 18 Oct 2010 07:40:42 +0000 fhr 1953@http://forums.flowingdata.com/ <p>Quantitative analyst, reporting to the Director of Operations.<br /> Location London; hours of work negotiable<br /> Fixed term position.</p> <p>This position is intended to add intellectual and quantitative firepower to our existing large-scale online advertising programme. The analyst will be responsible for:<br /> • Reviewing and either confirming or challenging our current theories about the levers of profit in online advertising<br /> • Data-mining our historical campaign information and using their deep statistical knowledge to extract new, testable ideas about highly targeted advertising strategies<br /> • Working with the existing team to translate these insights into action and profit</p> <p>Person specification</p> <p>1) Essential:<br /> - Upper second degree or equivalent in a quantitative subject (e.g. mathematics, physics, computer science)<br /> - Must be able to work independently<br /> - Detail-oriented<br /> - Very proficient in MS Excel<br /> - Experience with at least one common statistics and analysis package (e.g. R, SAS, MatLab)</p> <p>2) Desirable:<br /> - MA or PhD involving statistical analysis<br /> - Experience with relational databases and basic SQL<br /> - Experience with online advertising</p> <p>For more information please email francine[dot]bennett[at]ask.com </p> http://forums.flowingdata.com/topic/stanford-statistics-is-hiring#post-1946 Wed, 13 Oct 2010 15:41:51 +0000 nathany 1946@http://forums.flowingdata.com/ <p>Graduating soon? </p> <blockquote><p>The Statistics Department seeks applicants for a tenure track position at the Assistant Professor level. The position is to begin September, 2011. A doctorate in statistics or a related field (e.g., computer science, mathematics, engineering) is required. Applicants should have demonstrated strong research abilities, and will be expected to teach courses at both the undergraduate and graduate level. Applications are sought from all areas of statistics, both applied and theoretical. </p></blockquote> <p><a href="http://www-stat.stanford.edu/juniorfaculty.html" rel="nofollow">http://www-stat.stanford.edu/juniorfaculty.html</a> </p> http://forums.flowingdata.com/topic/data-science#post-1941 Tue, 12 Oct 2010 12:53:55 +0000 nathany 1941@http://forums.flowingdata.com/ <p>Computer science. Simply because I think it's easier to be a CS major and take stat courses than it is for a Stat major to take CS courses.</p> <p>Oh, and don't forget the design part. </p> http://forums.flowingdata.com/topic/data-science#post-1939 Tue, 12 Oct 2010 10:26:10 +0000 amathew 1939@http://forums.flowingdata.com/ <p>There are many skills that go into making a data scientists. In regards to education, it often is computer science and statistics. However, I was wondering what which undergraduate major would provide better training for becoming a data scientist? Computer science, applied math, statistics, etc. </p> http://forums.flowingdata.com/topic/annual-data-expo-2011-deep-water-horizon-water-spill#post-1938 Tue, 12 Oct 2010 01:57:18 +0000 nathany 1938@http://forums.flowingdata.com/ <p>do you have what it takes? this year's data expo dataset has just been announced. Explore it, understand, visualize it. Win.</p> <p><a href="http://streaming.stat.iastate.edu/dataexpo/2011/" rel="nofollow">http://streaming.stat.iastate.edu/dataexpo/2011/</a> </p> http://forums.flowingdata.com/topic/another-education-question#post-1921 Fri, 01 Oct 2010 12:56:15 +0000 amathew 1921@http://forums.flowingdata.com/ <p>This is the OP again. </p> <p>I should have mentioned that I am Really not interested in being one of those career students. Originally, my plan was to be done with my accademic career by age 25, which I turn today. So while I'm willing to return to school for another degree, I'm hoping to minimize the time in school. </p> http://forums.flowingdata.com/topic/another-education-question#post-1920 Fri, 01 Oct 2010 12:24:30 +0000 amathew 1920@http://forums.flowingdata.com/ <p>I recently complete my Master of Arts degree in a non-quantitative program. As an undergrad, I've taken nothing above one semester of calculas. During grad school, I had three statistical methods courses on OLS, Logit, Probit, etc.</p> <p>I'm now running a small analytics company which helps small business clients with data analysis and visualization. Much of what I am asked to do is simple stuff, univariate distributions, OLS, etc. However, I've been thinking about returning to school to furture my education. I could pursue a BS in statistics, or I've considered the iSchool at Berkely.</p> <p>Does anyone have any suggestions regarding the next step in my education? Should I go back to school? What program? </p> http://forums.flowingdata.com/topic/public-opinion-visualization#post-1899 Wed, 22 Sep 2010 19:55:23 +0000 amathew 1899@http://forums.flowingdata.com/ <p>Does anyone know of any visualization concerning public opinion?</p> <p>Statistically, I would generally run a logit, probit, etc. and present the numeric findings and also show the predicted probabilities for important variables using the Zelig package in R. </p> <p>However, I'm looking for other examples of how people have presented public opinion data in a meaningful way...anyone??? </p> http://forums.flowingdata.com/topic/tracking-music-play-statistics-via-itunes#post-1733 Fri, 14 May 2010 10:32:44 +0000 carolinebeavon 1733@http://forums.flowingdata.com/ <p>Hey</p> <p>This sounds like a great project.</p> <p>I tackled the "genre/plays etc" question for my MA Online Journalism - using Manyeyes. At the time I had been looking into ways to capture live data from my iTunes library and I drew a blank. (although granted I am new to data visualization). </p> <p>In the end I resorted to a "snapshot" of my listening and analyzed the results, and the anomalies it threw up.</p> <p><a href="http://carolinebeavon.com/2010/03/18/looks-like-im-not-into-metal-any-more-toto/" rel="nofollow">http://carolinebeavon.com/2010/03/18/looks-like-im-not-into-metal-any-more-toto/</a></p> <p>I wonder if there is any opportunities now Spotify have started letting users import and sync their iTunes library/playlists - there must be some kind of data transfer going on - maybe it is something you could exploit?</p> <p>I'll be interested to see how your project develops - and good luck with it!</p> <p>Caroline </p> http://forums.flowingdata.com/topic/masters-degree-in-statistics#post-1674 Wed, 07 Apr 2010 12:28:50 +0000 formeraa 1674@http://forums.flowingdata.com/ <p>That is a decent ranking, but favors the more theoretical programs. If Rejectthenull is looking for applied programs, I suggest looking at Ohio State's Applied Stats M.S. program. In addition, <a href="http://www.kdnuggets.com" rel="nofollow">http://www.kdnuggets.com</a> has a listing of data mining-oriented programs. </p> http://forums.flowingdata.com/topic/masters-degree-in-statistics#post-1432 Mon, 18 Jan 2010 18:45:46 +0000 nathany 1432@http://forums.flowingdata.com/ <p>There's this, but it's kind of fuzzy ranking, I think:</p> <p><a href="http://graduate-school.phds.org/rankings/statistics/rank?w26=2&amp;w2=-4&amp;w17=5&amp;w3=4&amp;w8=5&amp;w1=5&amp;w18=-2&amp;w9=3&amp;w24=-2&amp;w10=3&amp;w25=2&amp;w7=5" rel="nofollow">http://graduate-school.phds.org/rankings/statistics/rank?w26=2&amp;w2=-4&amp;w17=5&amp;w3=4&amp;w8=5&amp;w1=5&amp;w18=-2&amp;w9=3&amp;w24=-2&amp;w10=3&amp;w25=2&amp;w7=5</a></p> <p>And of course, US News &amp; World Report has their annual rankings. </p> http://forums.flowingdata.com/topic/masters-degree-in-statistics#post-1431 Mon, 18 Jan 2010 18:43:07 +0000 Rejectthenull 1431@http://forums.flowingdata.com/ <p>Are there sources that compare/rank graduate schools with respect to the master's degree programs in statistics? I am thinking about comparing/ranking schools on dimensions like applied vs. theory, emphasis on programming, emphasis on data visualization, emphasis on web analytics, emphasis on teaching consulting skills, etc. </p> http://forums.flowingdata.com/topic/tracking-music-play-statistics-via-itunes#post-1283 Mon, 16 Nov 2009 16:33:23 +0000 nathany 1283@http://forums.flowingdata.com/ <p>i've thought about this too some. the tough part is getting patterns over time because, like you said, the iTunes XML only gives aggregates.</p> <p>the quick and dirty way i would go about it would be to write a script that parses your iTunes XML each day, and then sends the data over your.flowingdata using the Twitter API. Then you could run the script once a day or whatever.</p> <p>Or maybe you could try something with last.fm?</p> <p>@<a href="http://twitter.com/nairb774" rel="nofollow">nair77b</a> makes use of the last.fm API and then logs listening with YFD:</p> <p><a href="http://your.flowingdata.com/nairb774/page/291/" rel="nofollow">http://your.flowingdata.com/nairb774/page/291/</a> </p> http://forums.flowingdata.com/topic/tracking-music-play-statistics-via-itunes#post-1271 Wed, 11 Nov 2009 04:24:20 +0000 artik 1271@http://forums.flowingdata.com/ <p>I'm looking to analyze the music I listen to over a period of time. Not so much stats on artists, genres, ratings or song-lengths, but more as an exercise to discern a pattern in my listening preferences over time (say beginning of week vs end of week, particular times of the day, music to accompany my daily commute).</p> <p>I did find an article that had a few promising links (<a href="http://www.tunequest.org/a-look-at-itunes-statistics-options/20060904/" rel="nofollow">http://www.tunequest.org/a-look-at-itunes-statistics-options/20060904/</a>). </p> <p>The problem with most of the software linked to in the article was that it did not give me statistics or analysis over time. This could be a problem with the depth of data available in the iTunes XML file, but I'd be willing to sync my iPhone/iPod every day for a month or two, if that's what is needed. </p> <p>I considered using yfd, but I'm concerned that given the amount of time I listen to music at work, it might cause severe disruptions in my work day if I had to tweet every time I make a conscious song/artist/genre choice.</p> <p>Your thoughts? </p> http://forums.flowingdata.com/topic/postdoc-opportunity-at-harvard-dept-of-statistics#post-707 Wed, 27 May 2009 22:02:30 +0000 nathany 707@http://forums.flowingdata.com/ <p><a href="http://www.stat.columbia.edu/~cook/movabletype/archives/2009/05/postdoc_opportu_2.html" rel="nofollow">http://www.stat.columbia.edu/~cook/movabletype/archives/2009/05/postdoc_opportu_2.html</a></p> <blockquote><p>We have two postdoctoral fellowships available for up-to three years, with competitive salary and travel support. The two postdocs are expected to contribute to active research projects, including...</p></blockquote>