måndag 26 december 2016

What is Metadata and why is it as important as the data itself?

From Dataconomy
Metadata. You may have heard the term before, and may have asked yourself either “what is metadata” or “why is it as important as data?” This article will be an attempt to clear up those two subjects. As this can often be quite dense, let’s jump right in!

fredag 23 december 2016

Flyttlassen fortsätter från storstäderna

befolkningsforandring_stockholm_2000-2015
Från Statisticon Befolkningsprognoser:

– inrikes flyttare väljer pendlingsnära kommuner

I förra inlägget beskrevs hur folkmängden i svenska landsbygdskommuner totalt sett växer. Vi tog också upp frågan om urbanisering och kunde konstatera att svenska städer förvisso växer fortare än landsbygdskommuner – men inte på bekostnad av dessa. De tre storstäderna Stockholm, Göteborg och Malmö växer främst på grund av utrikes inflyttning och att det föds många barn. I detta inlägg belyser vi frågan om inrikes flyttströmmar med storstaden Stockholm och pendlingsnära kommuner som exempel. Underlaget bygger på registrerade flyttningar under perioden januari till september 2016 och kommer från SCB:s befolkningsregister.

Bilden visar på vilket sätt Stockholms stad har vuxit under åren 2000-2015. Den totala folkmängden har under perioden ökat med 180 000 personer med en inledande blygsam ökning. Under åren 2008-2010 var ökningen som störst, för att sedan minska fram till år 2015.

Läs mer....

Hur fördelas BNP-tillväxten bland hushållen?

 image
Från Ekonomistas
Ett av de största problemen med de senaste decenniernas inkomstfördelningsanalyser är att de inte täckt in alla inkomster i ekonomin. Exempelvis tillfaller en stor del av arbetsgivaravgiften faktiskt löntagaren i form av försäkringsinkomster vid sjukdom, arbetslöshet eller ålderdom. Merparten av företagens vinster återinvesteras och syns aldrig i ägarnas inkomstdeklarationer. I en ny, banbrytande studie visar fransmännen Thomas Piketty, Emmanuel Saez och Gabriel Zucman att endast 60 procent av nationalinkomsten i USA syns i hushållens taxerade inkomster. Genom att fördela de resterande 40 procenten skapar forskarna en databas för USA där hela den makroekonomiska tillväxten under 1900-talet fördelas på de amerikanska hushållen.

onsdag 21 december 2016

Pokemon GO Has Some Massive Statistics, So What

From Games : iTech Post
Niantic has released the latest Pokemon GO update and along with that, the company also revealed some notable statistics for the game. However, those statistics seemed unimportant compared to the update.

Who said it?

From StatsLife
The Royal Statistical Society's Christmas Quiz is highly regarded for the level of challenge it provides. But if you are looking for something a little less taxing at this time of the year, the Significance Quotes Quiz has you covered. Our thanks to Jim Norton for putting the questions together. Have fun, thank you for reading, and we look forward to welcoming you back in 2017.

Read more.....

What Big Data and Predictive Analytics Missed in 2016

From Datanami Lessons Learned
In this era of the software-driven business, we’re told “data is the new oil”, and that predictive analytics and machine intelligence will extract actionable insights from this valuable resource and revolutionize the world as we know it. Yet, 2016 brought three highly visible failures in this predictive view of the world: the UK’s Brexit plebiscite; the Colombian referendum on FARC; and finally, the U.S. presidential election. What did these scenarios have in common? They all dealt with human behavior. This got me thinking that there might be lessons to be learned that are relevant to analytics.

Europeans greatly overestimate Muslim population, poll shows

People in Antwerp hold a beach party this summer to protest at the burkini ban over the border in France.
Photograph: Alamy
From The Guardian
Members of the public in European states including France, Belgium, Germany and the UK greatly overestimate their country’s Muslim population and the rate at which it is growing.

An Ipsos Mori survey that measured the gap between public perception and reality in 40 countries in 2016 found French respondents were by far the most likely to overstate their country’s current and projected Muslim population.

"People living in extreme poverty has decreased"

Gapminder's new Public Educator Helena, shows how Extreme Poverty went from the global default to an exception. http://buff.ly/2h6XNo5



Helena Nordenstedt, Assistant Professor in Global Health, Karolinska Institutet holds a presentation at the conference "Global Health in Conflict, Poverty and Fragility" on May 30th 2016 in Stockholm, organised by the think tank Global Utmaning.

The conference was a platform for dialogue and knowledge dissemination, with the aim to increase awareness about the health challenges in fragile and conflict-affected environments, and to reach an understanding of what actions need to be taken to ensure that fragile and conflict-affected states are not left behind in achieving the SDGs.

Global Utmaning (Global Challenge) is an independent think tank that promotes long-term solutions to crises and challenges in the ecological, economic and social systems through collaboration between research, business, politics and civil society.

http://www.globalutmaning.se #GUhealth16 #Agenda2030

The biggest stats lesson of 2016

From STATS  
What is the big statistical lesson of 2016? Here at STATS.org, we believe 2016’s major message is that statistical issues should be reported clearly and frequently to avoid miscommunication to lay audiences. This message was highlighted by the mismatch between the 2016 presidential election predictions and outcome.

What if you flip a coin twice, and both times it lands on “heads.” Does the coin seem rigged? The chance of two heads over two coin flips is 25 percent—less than the 29 percent chance that ESPN’s FiveThirtyEight gave Donald Trump for winning the presidency [1]. Although this probability was smaller than the 71 percent probability assigned to a Hillary Clinton win based on the final FiveThirtyEight pre-election forecast, the Trump win wasn’t as shocking an outcome as some journalists reported.

tisdag 20 december 2016

The Role of Statistics in Business Decision Making

From Talend
The use of statistics in business can be traced back hundreds of years. As early as 744 AD, statistics were used by Gerald of Wales to complete the first population census of Wales (1). It wasn’t long before merchants realized that statistics could be used to measure and quantify trade. The first record of this was in Florence. It was recorded in Giovanni Villani’s “Nuova Cronica”, in 1346 (1). Moreover, statistical methods were further adopted to help drive quality and in doing so helped contribute to the advancement of statistics itself. In 1504, William Sealy Gosset, chief brewer for Guinness in Dublin, devised the t-test (2) to measure consistency between batches of stout (1).

Read more.....

Räkna med bortfall

51tnwtuajil-_sx352_bo1204203200_
Talande omslagsbild, från Ineke A.L. Stoops bok 
Från Politologerna: Lucka #20: Räkna med bortfall 
Urvalsundersökningar är fantastiska – genom att fråga ett slumpmässigt urval om några tusen personer kan vi uttala oss uppfattningar i en befolkning bestående av miljontals människor. Det finns visserligen en osäkerhet i de resultat vi når, men den osäkerheten kan preciseras och kvantifieras. Detta tack vare den statistiska teori som ligger som grund för urvalsundersökningar och därmed även för en stor del av den samhällsvetenskapliga forskningen (se tidigare inlägg för utförligare historik).

Tyvärr verkar inte alla hysa varma känslor för sådana undersökningar; det finns de personer som vägrar att delta när opinionsinstituten hör av sig. På så vis blir de en del av bortfallet. I de klassiska böckerna om statistisk urvalsmetodik ägnades inte många rader åt bortfall, men i takt med att färre svarar på undersökningar har bortfallsproblematiken blivit ett allt mer uppmärksammat forskningsområde (se t.ex. en färsk avhandling i statistik av Minna Genbäck, Umeå universitet). I Sverige har frågan de senaste åren lyfts fram på nyhetsplats i massmedia (se t.ex. DN) och även, lite oväntat, varit föremål för debatt på landets stora debattsidor (se t.ex här med repliker här och här, här, eller här med repliker här och här).

torsdag 15 december 2016

Three minutes with Hans Rosling will change your mind about the world

Photo: Jörgen Hildebrandt
From Nature News & Comment 
He has influenced leaders from Melinda Gates to Fidel Castro. Now, he is on a mission to save people from their preconceived ideas.

Hans Rosling knew never to flee from men wielding machetes. “The risk is higher if you run than if you face them,” he says. So, in 1989, when an angry mob confronted him at the field laboratory he had set up in what is now the Democratic Republic of the Congo, Rosling tried to appear calm. “I thought, ‘I need to use the resources I have, and I am good at talking’.”

fredag 9 december 2016

More on Statistics vs Data Science

Information Management News
From Information Management Blogs:
I met up with an old stats grad school friend the other day. When last we got together a few years ago, he went on a rant about “data science”, suggesting the term's nothing more than a pretentious new moniker for the same statistical work he's been doing for 35 years. I disagreed, noting a substantial evolution from our early statistics days in the breadth of problems, especially involving computation, we address today. I guess his thinking about the statistics-data science divide was akin to FiveThirtyEight's Nate Silver; mine was more like statistician Andrew Gelman.

I was a bit surprised to note my friend had mellowed little in his statistical thinking. He did acknowledge that predictive modeling from traditional statistics serves a different purpose than the machine learning prominent in business today – and, more importantly, that both types must now be part of the modeler's arsenal.

Konsolideringsbidrag till ung statistiker

Från Maria Karlsson, Umeå universitet
Vetenskapsrådet (VR) har för första gången delat ut s.k. konsolideringsbidrag till "de mest framstående yngre forskarna för att ge dem möjlighet att konsolidera sin forskningssatsning och vidga och bredda sin verksamhet”. 20 bidrag delades ut och ett av bidragen gick till docent Ingeborg Waernbaum, Enheten för statistik vid Umeå Universitet för projektet Utveckling av statistiska metoder för att analysera kausala effekter med populationsbaserade register. Ingeborg Waernbaum får totalt 12 miljoner (2 milj/år för år 2017-2022).

Mer info Vetenskapsrådet och pressmeddelande Umeå universitet


onsdag 7 december 2016

From UNESCO’s descriptive statistics to deductive Big Data: the role of human annotation in quantification processes

From IEDES, UMR 201, Université Paris 1 Panthéon-Sorbonne  
Analysis of the characteristics and activities of UNESCO’s statistical personnel indicates: (i) the intertwined nature of practices involved in data-production processes and of the knowledge this requires; (ii) the importance of human-annotation to production and maintenance of databases; (iii) the shift from descriptive statistics to statistical inference, all in the context of structured data. These findings help to define four recent quantification trends. As massive unstructured data takes over: (i) there is a greater reliance on machine learning and modeling; (ii) the use of supervised learning implies increasingly complex and diverse data annotation–which may modify the roleplayed by the social sciences; (iii) in unsupervised learning, based on non-annotated data, the role of statistical models is enhanced; (iv) in both cases inductive and deductive approaches may be of use. These trends are taken here to be represented by the expression “deductive quantification”.

tisdag 6 december 2016

Pisa Results - Compare your country by OECD

Compare your country by OECD:

The headline indicator for the three subject areas: science, mathematics and reading. Average performance refers to all 15-year-old students in a country/economy regardless of the school type and grade attended. Small differences between countries and over time may be statistically insignificant.

More info.....
and here (pdf)

Pisa resultat - Sök på Google

Pisa resultat - Sök på Google:
Artikelbild för pisa resultat från Aftonbladet

Glädjebeskedet: Svenska elever höjer sina resultat i alla ämnen
Aftonbladet-41 minuter sedan
Glädjebeskedet: Svenska elever höjer sina resultat i alla ämnen ... Nu är det alltså dags igen, ett nytt Pisa-resultat ska presenteras av ...

måndag 5 december 2016

Winter Conference in Statistics 2017 - registration is now open

Åre, Sweden, 12-16 March, 2017
From The organising committee
The Winter Conference in Statistics in Åre, 12-16 March, 2017, with topic Statistical analyses of big and high dimensional data – approached by Generalized Additive Models and Functional Data Analysis is now open for registration. 

fredag 2 december 2016

Kärleken till svenska data

big_data
Foto: Luckey_sun.
Från Politologerna: Kärleken till svenska data 
”Gotta love that Swedish administrative data!” Den kärleksförklaringen yttrades av Sarah Kliff i ett avsnitt tidigare i år av den utmärkta amerikanska podcasten the Weeds från Vox. Det finns många podcasts som fokuserar på amerikansk politik, men det som särskiljer the Weeds att är de dyker djupt ned i olika politiska förslag och diskuterar vad olika reformer skulle få för följder. I varje avsnitt presenterar de dessutom ett working paper som innehåller nya, spännande resultat från den samhällsvetenskapliga forskningen. Ofta tar de upp något working paper från the National Bureau of Economic Research (NBER), så även i det avsnitt då svenska administrativa data hyllades. Det papper som avhandlades studerade vilka effekter stress hos en gravid kvinna får på fostrets framtida liv.

Läs mer....

Data science: Unpacking data visualisation

From Data science: Unpacking data visualisation - YouTube

Data visualisation and the newsroom
Alan Smith,
Financial Times