stories   61857

« earlier    

Karl Pearson and the Chi-squared Test
Pearson's paper of 1900 introduced what subsequently became known as the chi-squared test of goodness of fit. The terminology and allusions of 80 years ago create a barrier for the modern reader, who finds that the interpretation of Pearson's test procedure and the assessment of what he achieved are less than straightforward, notwithstanding the technical advances made since then. An attempt is made here to surmount these difficulties by exploring Pearson's relevant activities during the first decade of his statistical career, and by describing the work by his contemporaries and predecessors which seem to have influenced his approach to the problem. Not all the questions are answered, and others remain for further study.

original paper:

How did Karl Pearson come up with the chi-squared statistic?:
He proceeds by working with the multivariate normal, and the chi-square arises as a sum of squared standardized normal variates.

You can see from the discussion on p160-161 he's clearly discussing applying the test to multinomial distributed data (I don't think he uses that term anywhere). He apparently understands the approximate multivariate normality of the multinomial (certainly he knows the margins are approximately normal - that's a very old result - and knows the means, variances and covariances, since they're stated in the paper); my guess is that most of that stuff is already old hat by 1900. (Note that the chi-squared distribution itself dates back to work by Helmert in the mid-1870s.)

Then by the bottom of p163 he derives a chi-square statistic as "a measure of goodness of fit" (the statistic itself appears in the exponent of the multivariate normal approximation).

He then goes on to discuss how to evaluate the p-value*, and then he correctly gives the upper tail area of a χ212χ122 beyond 43.87 as 0.000016. [You should keep in mind, however, that he didn't correctly understand how to adjust degrees of freedom for parameter estimation at that stage, so some of the examples in his papers use too high a d.f.]
nibble  papers  acm  stats  hypothesis-testing  methodology  history  mostly-modern  pre-ww2  old-anglo  giants  science  the-trenches  stories  multi  q-n-a  overflow  explanation  summary  innovation  discovery  distribution  degrees-of-freedom  limits 
yesterday by nhaliday
Anthropocene, Capitalocene, Plantationocene, Chthulucene: Making Kin
One way to live and die well as mortal critters in the Chthulucene is to join forces to reconstitute refuges, to make possible partial and robust biological-cultural-political-technological recuperation and recomposition, which must include mourning irreversible losses. Thom van Dooren and Vinciane Despret taught me that.10 There are so many losses already, and there will be many more. Renewed generative flourishing cannot grow from myths of immortality or failure to become-with the dead and the extinct. There is a lot of work for Orson Scott Card's Speaker for the Dead.11 And even more for Ursula LeGuin's worlding in Always Coming Home.

I am a compost-ist, not a posthuman-ist: we are all compost, not posthuman. The boundary that is the Anthropocene/Capitalocene means many things, including that immense irreversible destruction is really in train, not only for the 11 billion or so people who will be on earth near the end of the 21st century, but for myriads of other critters too. (The incomprehensible but sober number of around 11 billion will only hold if current worldwide birth rates of human babies remain low; if they rise again, all bets are off.) The edge of extinction is not just a metaphor; system collapse is not a thriller. Ask any refugee of any species.
donnaharaway  anthropocene  compost  stories 
2 days ago by oddhack
Electronic Texts of H.P. Lovecraft’s Works
Official website - All stories as electronic text as far as I know.
Stories  Tales  Texts  Howard  Phillips  Lovecraft  Complete 
6 days ago by Asselberghs
How Workday Bridges the Gap between Amazon and OpenStack
"“What I will strongly recommend is to build something,” Magana advised. “I keep calling it ‘middleware,’ but something that abstracts the back-end platform, your pipeline system, from the way to deliver your application artifact to production. I strongly recommend that.”

It is nothing like Aporeto’s CEO described it: succumbing to the relentless will of a monopolistic beast with its sights set on world domination. If anything, Workday’s strategy is a do-it-yourself program for making global public cloud deployment viable for an organization that is not, and may not ever be, rooted there to begin with."
case  study  technology  top  stories  amazon  ec2  container  service  web  services  aws  re:invent  2016  chef  continuous  integration  and  deployment  gerrit  jenkins  middleware 
6 days ago by jonerp

« earlier    

related tags

2016  2017  academia  accessibility  accuracy  acm  advertising  age-generation  agile  airbnb  alcoholism  allan  amazon  america  analysis  and  anthology  anthropocene  anthropology  anxiety  applauded  apple  architecture  arms  article  assessment  asynchrony  attention  author  autism  aviation  aviationweek  awesome  aws  babies  backup  bbc  beeld  behavioral-econ  biases  biography  bootstrapping  bosh  bowie  brca1  bugs  bullshit  bullying  business  businessmodels  buttercms  by  caltech  canon  case  cassandra  chart  chef  children  class  climatechange  cloud  cocktail  cog-psych  comics  communication  community  comparison  complete  compost  comprehension  computers  concentration  constructivism  container  continuous  contributed  copyright  cracker-econ  craigslist  creativity  crime  criticism  cryptocurrency  cv  data  data_management  davidbowie  decision-making  degrees-of-freedom  delivery  democracy  demographics  deployment  depression  design  developers  devops  digital  disaster  disasters  discovery  discrimination  discussion  distribution  diversity  diy  dohle  domain  donnaharaway  ec2  eclipse  economics  econotariat  edgar  education  el  embodied-pack  embodied  empirical  energy  environment  equilibrium  esb6  essay  essays  estimate  ethanol  events  evidence  explanation  facebook  fantasy  fiction  fighting  film  finance  firewire  food  football  for  foundry  fowler  francfort  freedom  funny  futurestack  ge  gender-diff  gender  general  gerrit  giants  government  guidelines  habits  harassment  health  healthcare  help  heroku  history  horror  how-to  howard  howto  hypothesis-testing  ifttt  impetus  innovation  instagram  integration  interdisciplinary  interface  internet  interviews  intrusion  ios  italy  jenkins  joe.dimaggio  journalism  kids  labyrinth  land  language  launch  law  lens  life  limits  lists  literature  longform  lovecraft  mac  macos  mainstream  manufacturing  marketing  markets  martin  mcdonalds  measurement  medium  mentalhealth  methodology  methods  meyer  microservices  middleware  mike  money  mostly-modern  movies  multi  narrative  narrativecharts  new  news  nibble  nonfiction  objektbuch  observation  old-anglo  on  op-ed  org:econlib  oriented  overflow  papers  parenting  patientexperience  peace-violence  people  perseverance  phillips  poe  politics  poll  pre-ww2  predictions  predix  prh  productivity  programming  psychology  q-n-a  qualitative  quality  quietude  r.a.lafferty  race  rape  re:invent  read  readalouds  realness  recycling  reflection  regularizer  religion  replication  research-program  research  resilience  restraint  running  sarahendren  science  scifi  service  services  sex  sharing  short  social-science  social  socialanxietydisorder  software  sondages  space  spanish  specific  sporfle  stats  stillness  storage  story  storycurves  storytelling  stratchery  street-fighting  study  summary  survey  sysops  tales  tax  techniques  technology  terror  testing  texts  the-trenches  the.moth  tips  top  totaleclipse  touchcast  track-record  transparency  trivia  tv  twitter  uk  user  userstories  video  watching  web  weinstein  witness  work  writer  writing  xml  yorker  🎩 

Copy this bookmark: