Category Archives: statistics

Statistics

Since it deals with data, whatever the data be, research in statistics is applicable to any discipline. While a grounding in a subject will help, it is possible to imagine a ‘pure statistician’ who can move across different domains and offer his expertise. There is no subject no problem that cannot be analysed with the help of statistics to find appropriate solutions. As long as there is information and data, there will be need for statistics. While many students are currently fascinated with data analytics or data science, they could well do Masters or PhD in statistics.

The Dedman College of Humanities and Sciences says that “The explosion of data from devices such as sensors, cell phones, and medical instruments, as well as from business processes, surveys, and social media has created an accelerating demand for specialists who are trained in data collection and analysis.  As a result, the field of statistics will be increasingly more fundamental to academia, business, and government in order to accommodate the escalating dependence on data-driven decisions” (http://www.smu.edu/Dedman/Academics/Departments/Statistics). The University of Hong Kong has a Statistical Advice Centre for Students which offers training courses in statistics (https://www.gradsch.hku.hk/gradsch/current-students/coursework-information/supporting-courses-services/statistical-advice-centre-for-students-stacs).

In what follows, we offer a peek into the many possibilities within statistics.

Let us first mention that probability, time series and multiple regression analyses are the most important topics that anyone needs to master if they wished to take up statistical analysis. Unfortunately, many students enter into areas such as analytics, data science or big data without a grounding in any of these three. Students often confuse familiarity with a statistical software with knowledge of statistics.

An article titled ‘Samples and Statistics: Distinguishing Populations of Hot Jupiters in a Growing Dataset’ describes a “new study examines how statistics can be used to characterize large amounts of data — for instance, distinguishing between populations of hot Jupiters”, using Bayesian methods (http://aasnova.org/2017/04/18/samples-and-statistics-distinguishing-populations-of-hot-jupiters-in-a-growing-dataset/). One of the most sought after areas of research is artificial intelligence together with statistics. Statistical relational artificial intelligence is a new area of research about which you may find more at https://www.frontiersin.org/research-topics/5640/statistical-relational-artificial-intelligence. An equally promising area is nonparametric statistical research which is used when knowledge is not precise enough to permit parametric methods. However, even within this field, there is an enormous range of research as can be seen at https://www.maths.unsw.edu.au/research/nonparametric-statistics.

Bayesian methods is a field that continues to be explore in the most diverse of contexts and subjects.  One of the areas where Bayesian methods are pursued is in public health, where we also encounter some of the most intense discussions of a methodological nature (http://www.bu.edu/sph/about/departments/biostatistics/research/methodological-research/).

The Advocacy action Center does a lot of statistical analyses in relation to the collection industry and the role it plays in debt and the economy (https://www.acainternational.org/advocacy/industry-research-statistics). The Pew Research Center uses statistics to analyse public opinion especially as it relates to politics and science issues understood within the democrat-republican framework (http://www.pewinternet.org/2015/07/01/americans-politics-and-science-issues/). The Society for the Study of Addiction puts together useful statistics on the prevalence and patterns of alcohol, drug and tobacco use (https://www.addiction-ssa.org/commentary/where-to-find-useful-statistics-and-information).

A relatively new area of research is the development and application of spatio-temporal statistics, which involves many challenging questions in theory, computation, simulation, and application (http://webs.wichita.edu/?u=mathsci&p=/mathstat/research/statistics/). Research in statistics at IIT Kanpur is a good overview of what is possible in such work (https://www.iitk.ac.in/math/research-areas-in-statistics). The University of Pennsylvania, Wharton, has put up a listing of research in statistics (https://statistics.wharton.upenn.edu/research/research-listing/). Another resource which furnishes a window to what is possible is https://www.sciencedaily.com/news/computers_math/statistics/. There is an extremely useful article on ‘How to do statistical research’ by Terry Speed, head of bioinformatics at the Walter & Eliza Hall Institute of Medical Research in Melbourne, Australia, and an active emeritus statistics professor at the University of California, Berkeley (http://stattrak.amstat.org/2013/06/01/how-to-do-statistical-research/). Wiley’s online library too offers a good view of the topic (http://onlinelibrary.wiley.com/doi/10.1111/j.1751-5823.2007.00029.x/abstract).

As we said at the beginning, statistics can be applied to any field. Here is an interesting article on the use of statistics in sports (www.mdpi.com/2075-4663/5/4/87/pdf). Another example of the diversity in statistical research is https://www.bls.gov/opub/mlr/2017/article/pdf/benchmarking-the-current-employment-statistics-survey-perspectives-on-current-research.pdf. The research being done by the National Center on Elder Abuse is a pointer to the importance of statistical research (https://ncea.acl.gov/whatwedo/research/statistics.html). The US Department of Education also undertakes statistical research relevant to education (https://ed.gov/rschstat/landing.jhtml?src=pn). The National Science Foundation gathers a lot of data to study business and industrial R&D (https://www.nsf.gov/statistics/industry/). Another interesting institute is The International Society for Business and Industrial Statistics which “promotes applications, research, and best current practices in business and industrial statistics, facilitates technology transfer, and fosters communications among members and practitioners worldwide”, (http://www.isbis-isi.org/). Cancer Research and Biostatistics (CRAB) will appeal to those who would like to do something in the field of diagnosis and cure for cancer (https://www.crab.org/services-projects.html). If you are interested in market intelligence, you may look at companies such as IDC (https://www.idc.com/) or Gartner (https://www.gartner.com/technology/home.jsp) or many similar ones.

A researcher at the Stockholm University has researched on a better way to forecast growth in GDP (Gross Domestic Product), a crucial indicator of how well an economy is performing (http://www.statistics.su.se/english/research/time-series-analysis/current-researlch).  It is widely known that statistics is extensively used in investment analysis and markets be they stock, commodity, financial or alternative. Consider this research which seeks to throw light on financial crises (https://www.sciencedaily.com/releases/2017/11/171122103559.htm). The University of British Columbia, Vancouver Campus, has a focus on biology and meta-research in its research in statistics (https://www.stat.ubc.ca/research-areas). The Statistics Research Group at Cardiff University is engaged in research into the subject itself (https://www.cardiff.ac.uk/research/explore/research-units/statistics). Trinity College of Arts & Sciences, Duke, has a broader range of research interests (https://stat.duke.edu/research) moving across physics, biology, advertising, ecology & environment, economics, finance & policy and the subject itself, while also engaging in interdisciplinary collaborative research.

A good indicator of the use of statistics in healthcare comes from The Department of Health, Australian Government, (http://www.health.gov.au/internet/main/publishing.nsf/content/health-pubhlth-strateg-active-evidence.htm). The National Clinical Assessment Centre of NHS uses statistical analyses as an integral part of its work (http://www.ncas.nhs.uk/about-ncas/statistics/).

Combining statistics with economics and social sciences, the University of Ghana has an Institute of Statistical, Social and Economic Research in the College of Humanities (http://isser.edu.gh/). Among other research interests which is quite a range, The University of Kent engages in research in nonparametric statistics (https://www.kent.ac.uk/smsas/statistics/research/np.html). MIT’s Statistics and Data Science Center is engaged in Bayesian nonparametric statistics (https://stat.mit.edu/research/nonparametric-bayesian-statistics/). Leiden University offers a Masters in Statistical Science for life and Behavioural Sciences (https://www.universiteitleiden.nl/en/education/study-programmes/master/statistical-science-for-the-life-and-behavioural-sciences). Students with a physics background may consider statistical physics as an area of work as it offers a vast array of topics (https://www.nature.com/subjects/statistical-physics).

The Statistics for Industry in the UK is an interesting institution which combines research, consulting and training for industry with some open courses too (http://s4i.co.uk/). Cornell University Library has published a useful list of institutions which record statistics based on the view that “The best strategy for finding statistics is to identify the stakeholders (governments, companies, organizations) and do a thorough search of their publications and Internet pages” (http://guides.library.cornell.edu/c.php?g=31402&p=199848). If you are looking for statistical research on investment, take a look at the Investment Company Institute which claims it is the best source “for statistical data and research on investors and retirement plans in the investment company industry”, (https://www.ici.org/research).

Advertisements