parsimony   110

« earlier    

Against parsimony - Statistical Modeling, Causal Inference, and Social Science
Sometimes a simple model will outperform a more complex model . . . Nevertheless, I believe that deliberately limiting the complexity of the model is not fruitful when the problem is evidently complex. Instead, if a simple model is found that outperforms some particular complex model, the appropriate response is to define a different complex model that captures whatever aspect of the problem led to the simple model performing well.
statistics  parsimony  model-selection 
june 2018 by arsyed
Plague of Frogs | West Hunter
For a few years the herpetologists were concerned yet happy. Concerned, because many frog populations were crashing and some were going extinct. Happy, because confused puppies in Washington were giving them money, something that hardly ever happens to frogmen. The theory was that amphibians were ‘canaries in a coal mine’, uniquely sensitive to environmental degradation.


It took some time for herpetologists to admit that this chytrid fungus is the main culprit – some are still resisting. First, it was a lot like how doctors resisted Semmelweiss’ discoveries about the cause of puerperal fever – since doctors were the main method of transmission. How did this fungus get to the cloud forests of Costa Rica? On the boots of herpetologists, of course.

The second problem is Occam’s butterknife: even though this chytrid fungus is the main culprit, it’s just got to be more complicated than that. Even if it isn’t. People in the life sciences – biology and medicine – routinely reject simple hypotheses that do a good job of explaining the data for more complex hypotheses that don’t. College taught them to think – unwisely.
west-hunter  scitariat  reflection  stories  troll  lol  science  low-hanging  occam  parsimony  bio  medicine  meta:medicine  ability-competence  explanans  disease  parasites-microbiome  spreading  world  nature  environment  climate-change  hypochondria  academia  questions  epidemiology  incentives  interests 
february 2018 by nhaliday
Ptolemy's Model of the Solar System
It follows, from the above discussion, that the geocentric model of Ptolemy is equivalent to a heliocentric model in which the various planetary orbits are represented as eccentric circles, and in which the radius vector connecting a given planet to its corresponding equant revolves at a uniform rate. In fact, Ptolemy's model of planetary motion can be thought of as a version of Kepler's model which is accurate to first-order in the planetary eccentricities--see Cha. 4. According to the Ptolemaic scheme, from the point of view of the earth, the orbit of the sun is described by a single circular motion, whereas that of a planet is described by a combination of two circular motions. In reality, the single circular motion of the sun represents the (approximately) circular motion of the earth around the sun, whereas the two circular motions of a typical planet represent a combination of the planet's (approximately) circular motion around the sun, and the earth's motion around the sun. Incidentally, the popular myth that Ptolemy's scheme requires an absurdly large number of circles in order to fit the observational data to any degree of accuracy has no basis in fact. Actually, Ptolemy's model of the sun and the planets, which fits the data very well, only contains 12 circles (i.e., 6 deferents and 6 epicycles).
org:junk  org:edu  nibble  physics  space  mechanics  history  iron-age  mediterranean  the-classics  science  the-trenches  the-great-west-whale  giants  models  intricacy  parsimony 
september 2017 by nhaliday
All models are wrong - Wikipedia
Box repeated the aphorism in a paper that was published in the proceedings of a 1978 statistics workshop.[2] The paper contains a section entitled "All models are wrong but some are useful". The section is copied below.

Now it would be very remarkable if any system existing in the real world could be exactly represented by any simple model. However, cunningly chosen parsimonious models often do provide remarkably useful approximations. For example, the law PV = RT relating pressure P, volume V and temperature T of an "ideal" gas via a constant R is not exactly true for any real gas, but it frequently provides a useful approximation and furthermore its structure is informative since it springs from a physical view of the behavior of gas molecules.

For such a model there is no need to ask the question "Is the model true?". If "truth" is to be the "whole truth" the answer must be "No". The only question of interest is "Is the model illuminating and useful?".
thinking  metabuch  metameta  map-territory  models  accuracy  wire-guided  truth  philosophy  stats  data-science  methodology  lens  wiki  reference  complex-systems  occam  parsimony  science  nibble  hi-order-bits  info-dynamics  the-trenches  meta:science  physics  fluid  thermo  stat-mech  applicability-prereqs  theory-practice 
august 2017 by nhaliday
Kelly criterion - Wikipedia
In probability theory and intertemporal portfolio choice, the Kelly criterion, Kelly strategy, Kelly formula, or Kelly bet, is a formula used to determine the optimal size of a series of bets. In most gambling scenarios, and some investing scenarios under some simplifying assumptions, the Kelly strategy will do better than any essentially different strategy in the long run (that is, over a span of time in which the observed fraction of bets that are successful equals the probability that any given bet will be successful). It was described by J. L. Kelly, Jr, a researcher at Bell Labs, in 1956.[1] The practical use of the formula has been demonstrated.[2][3][4]

The Kelly Criterion is to bet a predetermined fraction of assets and can be counterintuitive. In one study,[5][6] each participant was given $25 and asked to bet on a coin that would land heads 60% of the time. Participants had 30 minutes to play, so could place about 300 bets, and the prizes were capped at $250. Behavior was far from optimal. "Remarkably, 28% of the participants went bust, and the average payout was just $91. Only 21% of the participants reached the maximum. 18 of the 61 participants bet everything on one toss, while two-thirds gambled on tails at some stage in the experiment." Using the Kelly criterion and based on the odds in the experiment, the right approach would be to bet 20% of the pot on each throw (see first example in Statement below). If losing, the size of the bet gets cut; if winning, the stake increases.
nibble  betting  investing  ORFE  acm  checklists  levers  probability  algorithms  wiki  reference  atoms  extrema  parsimony  tidbits  decision-theory  decision-making  street-fighting  mental-math  calculation 
august 2017 by nhaliday
A Lttle More Nuance
economics lowest, though still increasing (also most successful frankly, wonder why? :P):
In this view, the trajectories of the disciplines relative to one another are sharpened. I have to say that if you’d asked me ex ante to rank fields by nuance I would have come up with an ordering much like the one visible at the end of the trend lines. But it also seems that social science fields were not differentiated in this way until comparatively recently. Note that the negative trend line for Economics is relative not to the rate of nuance within field itself—which is going up, as it is everywhere—but rather with respect to the base rate. The trend line for Philosophy is also worth remarking on. It differs quite markedly from the others, as it has a very high nuance rate in the first few decades of the twentieth century, which then sharply declines, and rejoins the upward trend in the 1980s. I have not looked at the internal structure of this trend any further, but it is very tempting to read it as the post-WWI positivists bringing the hammer down on what they saw as nonsense in their own field. That’s putting it much too sharply, of course, but then again that’s partly why we’re here in the first place.
scitariat  social-science  sociology  philosophy  history  letters  psychology  sapiens  anthropology  polisci  economics  anglo  language  data  trends  pro-rata  visualization  mostly-modern  academia  intricacy  vague  parsimony  clarity  multi  twitter  social  commentary  jargon  pdf  study  essay  rhetoric  article  time-series  lexical 
june 2017 by nhaliday
Rheumatoid Arthritis | West Hunter
It causes characteristic changes in the bones.  Key point:  it is vanishingly rare in Old World skeletons before the 17th century.  Those changes, however, been seen in some pre-Columbian Amerindian skeletons [work by Bruce Rothschild].

The obvious explanation is that RA is caused by some pathogen that originated in the Americas and later spread to the rest of the world.  Like the French disease.
Everybody knows that the Amerindians were devastated by new infectious diseases after Columbus discovered America and made it stick. Smallpox, falciparum malaria, yellow fever, bubonic plague, cholera, measles, whooping cough, etc : by some estimates, the Amerindian population dropped by about 90%, worse than the Black Plague, which only killed off half of Europe. Naturally, you wonder what ailments the Americas exported to the rest of the world.

We know of two for sure. First, syphilis: the first known epidemic was in 1495, in Naples, during a French invasion. By 1520 it had reached Africa and China.

From the timing of the first epidemic, and the apparent newness of the disease, many have suspected that it was an import from the New World. Some, like Bartolome de las Casas, had direct knowledge: Las Casas was in Seville in 1493, his father and uncle sailed with Columbus on the second voyage, and he himself traveled to the New World in 1502, where he spent most of the rest of his life working with the Amerindians. Ruiz Diaz de Isla, a Spanish physician, reported treating some of Columbus’s crew for syphilis, and that he had observed its rapid spread in Barcelona.

I have seen someone object to this scenario, on the grounds that the two years after Columbus’s return surely couldn’t have been long enough to generate a major outbreak. I think maybe that guy doesn’t get out much. It has always looked plausible, considering paleopathological evidence (bone changes) and the timing of the first epidemic. Recent analysis shows that some American strains of pinta (a treponemal skin disease) are genetically closest to the venereal strains. I’d say the Colombian theory is pretty well established, at this point.

Interestingly, before the genetic evidence, this was one of the longest-running disputes among historians. As far as I can tell, part of the problem was (and is) that many in the social sciences routinely apply Ockham’s razor in reverse. Simple explanations are bad, even when they fit all the facts. You see this in medicine, too.


There are two other diseases that are suspected of originating in the Americas. The first is typhus, gaol fever, caused by a Rickettsial organism and usually spread by lice. Sometimes it recurs after many years, in a mild form called Brill’s disease, rather like chickenpox and shingles. This means that typhus is always waiting in the wings: if the world gets sufficiently messed up, it will reappear.

Typhus shows up most often in war, usually in cool countries. There is a claim that there was a clear epidemic in Granada in 1489, which would definitely predate Columbus, but descriptions of disease symptoms by premodern physicians are amazingly unreliable. The first really reliable description seems to have been by Fracastoro, in 1546 (according to Hans Zinsser in Rats, Lice, and History). The key hint is the existence of a very closely related organism in American flying squirrels.

Thinking about it, I have the impression that the legions of the Roman Republic didn’t have high casualties due to infectious disease, while that was the dominant cause of death in more recent European armies, up until the 20tth century. If smallpox, measles, syphilis, bubonic plague, perhaps typhus, simply hadn’t arrived yet, this makes sense. Falciparum malaria wasn’t much of a factor in northern Italy until Imperial times…

The second possibly American disease is rheumatoid arthritis. We don’t even know that it has an infectious cause – but we do know that it causes characteristic skeletal changes, and that no clear-cut pre-Columbian rheumatoid skeletons are known from the Old World, while a number have been found in the lower South. To me, this makes some infectious cause seem likely: it would very much be worth following this up with the latest molecular genetic methods.

American crops like maize and potatoes more than canceled the demographic impact of syphilis and typhus. But although the Old World produced more dangerous pathogens than the Americas, due to size, longer time depth of agriculture, and more domesticated animals, luck played a role, too. Something as virulent as smallpox or falciparum malaria could have existed in the Americas, and if it had, Europe would have been devastated.
Malaria came from Africa, probably. There are old primate versions. Smallpox, dunno: I have heard people suggest viral infections of cows and monkeys as ancestral. Measles is derived from rinderpest, probably less than two thousand years ago.

Falciparum malaria has been around for a while, but wasn’t found near Rome during the Republic. It seems to have gradually moved north in Italy during classical times, maybe because the range of the key mosquito species was increasing. By early medieval times it was a big problem around Rome.

Smallpox probably did not exist in classical Greece: there is no clear description in the literature of the time. It may have arrived in the Greco-Roman world in 165 AD, as the Antonine plague.

The Pathogenesis of Rheumatoid Arthritis:
In the Age of Discovery, Europeans were playing with fire. Every voyage of exploration risked bring back some new plague. From the New World, syphilis, probably typhus and rheumatoid arthritis. From India, cholera. HIV, recently, from Africa. Comparably important new pests attacking important crops and domesticated animals also arrived, such as grape phylloxera (which wiped out most of the vineyards of Europe) and potato blight ( an oomycete or ‘water mold’, from central Mexico).

If one of those plagues had been as potent as smallpox or falciparum malaria, you probably wouldn’t be reading this.
west-hunter  scitariat  discussion  ideas  speculation  critique  disease  parasites-microbiome  usa  age-of-discovery  europe  embodied  history  early-modern  multi  spreading  random  counterfactual  🌞  occam  parsimony  archaeology  cost-benefit  india  asia  africa  agriculture  uncertainty  outcome-risk  red-queen  epidemiology  thick-thin  pdf  piracy  study  article  survey  iron-age  the-classics  mediterranean  novelty  poast 
may 2017 by nhaliday
The Wordsmith Shares His Craft - WSJ
By Edward Kosner
May 17, 2017


By Harold Evans
Little, Brown, 408 pages, $27

Mr. Evans introduces a crisp curriculum of do’s and don’ts for the aspiring clear writer. He counsels the active voice over the passive, the parsimonious use of adjectives and the near banishment of adverbs. (Not as easily practiced as preached.) He also urges writers to cut fat, check their math, be specific, organize their material for clarity, accentuate the positive and never be boring.
clarity  editors  writing  book_reviews  books  parsimony  words  wordsmiths 
may 2017 by jerryking
Mean field theory - Wikipedia
In physics and probability theory, mean field theory (MFT also known as self-consistent field theory) studies the behavior of large and complex stochastic models by studying a simpler model. Such models consider a large number of small individual components which interact with each other. The effect of all the other individuals on any given individual is approximated by a single averaged effect, thus reducing a many-body problem to a one-body problem.
concept  atoms  models  physics  stat-mech  ising  approximation  parsimony  wiki  reference  nibble 
march 2017 by nhaliday
Information Processing: Epistasis vs additivity
On epistasis: why it is unimportant in polygenic directional selection:
- James F. Crow

The Evolution of Multilocus Systems Under Weak Selection:
- Thomas Nagylaki

Data and Theory Point to Mainly Additive Genetic Variance for Complex Traits:
The relative proportion of additive and non-additive variation for complex traits is important in evolutionary biology, medicine, and agriculture. We address a long-standing controversy and paradox about the contribution of non-additive genetic variation, namely that knowledge about biological pathways and gene networks imply that epistasis is important. Yet empirical data across a range of traits and species imply that most genetic variance is additive. We evaluate the evidence from empirical studies of genetic variance components and find that additive variance typically accounts for over half, and often close to 100%, of the total genetic variance. We present new theoretical results, based upon the distribution of allele frequencies under neutral and other population genetic models, that show why this is the case even if there are non-additive effects at the level of gene action. We conclude that interactions at the level of genes are not likely to generate much interaction at the level of variance.
hsu  scitariat  commentary  links  study  list  evolution  population-genetics  genetics  methodology  linearity  nonlinearity  comparison  scaling-up  nibble  lens  bounded-cognition  ideas  bio  occam  parsimony  🌞  summary  quotes  multi  org:nat  QTL  stylized-facts  article  explanans  sapiens  biodet  selection  variance-components  metabuch  thinking  models  data  deep-materialism  chart  behavioral-gen  evidence-based  empirical  mutation  spearhead  model-organism  bioinformatics  linear-models  math  magnitude  limits  physics  interdisciplinary  stat-mech 
february 2017 by nhaliday

« earlier    

related tags

@todo  ability-competence  academia  academic  accuracy  acm  acmtariat  action-oriented  aesthetics  africa  age-of-discovery  agile  agriculture  ai  algorithms  alt-inst  amar_bhidé  analysis  analysis_paralysis  analytical-holistic  andrew_gelman  anglo  anthropology  api  apollonian-dionysian  applicability-prereqs  approximation  archaeology  architecture  arms  article  arts  asia  atoms  axioms  backup  bayesian  beauty  behavioral-gen  best.practices  betting  biases  big-list  big-peeps  big-picture  bing  bio  biodet  bioinformatics  biology  biotech  blog  blogs  book_reviews  books  borges  bounded-cognition  broad-econ  browser  buses  c++  c:**  calculation  calendar  capital_efficiency  carcinisation  chart  checklists  chicago  child  chomsky  christianity  civilization  cladistics  clarity  cli  client  climate-change  cloud  cocktail  code  coding  collaborative  commentary  communism  community  comparison  competition  compiler  complex-systems  complexity  composition-decomposition  composition  compression  compsci  computational-complexity  concept  conceptual-vocab  concurrency  confusion  conquest-empire  consider:not-doing-it-in-that-order  consider:stress-testing  consilience  contradiction  contrarianism  convert  cosma_shalizi  cost-benefit  counterfactual  cpod  critique  crooked  cruft  css  culture  cybernetics  cycles  dangers-of-premature-optimization  data-analysis  data-science  data  datamining  dataviz  death  decision-making  decision-theory  deep-learning  deep-materialism  deep_structure  deficits  definition  dennett  design  detail-architecture  detail-oriented  development  discovery  discrete  discussion  disease  distributed  django  do-it-again  don_drummond  drugs  duplication  dynamic-homology  dynamical  early-modern  economics  economy  econotariat  editors  education  efficiency-is-on-the-table-now-i-see  efficiency  elite  email  embodied  emergent  emergent_properties  empirical  entrepreneur  environment  epidemiology  epistemic  epistemology  equilibrium  error  essay  europe  evidence-based  evidence  evolution  excellent  exegesis-hermeneutics  explanans  explanation  extrema  facebook  filetype:pdf  finance  flexibility  flexure  fluid  flux-stasis  food  forms  founders  fourier  framework  frontier  frugality  garett-jones  gender  genetic  genetics  geocoding  giants  gibbon  gis  good_enough  google  government  grammar  grammarthing  graph_rewriting  ground-up  growth-econ  gtd  hanson  hayek  hbr  health  hetero-advantage  heterodox  heuristic  hi-order-bits  high-quality  hinge  history  hmm  homo-hetero  hosting  howto  hsu  html  http  hustle  hypochondria  ideas  ideology  idk  ieee  image-processing  image-segmentation  impetus  incentives  india  inference  info-dynamics  infrastructure  init  insight  interdisciplinary  interests  intricacy  intuition  investing  iron-age  ising  jargon  javascript  jesus-christ-people  john-horton-conway  jorge_luis_borges  jquery  karl_smith  krugman  language  larry_wasserman  law  laws  learning  left-wing  len:long  len:short  lens  lesson-for-genetic-programming  lesswrong  letters  levers  lexical  library2.0  likelihood  limits  linear-models  linearity  linguistics  links  list  logic  lol  low-cost  low-hanging  machine-learning  macro  macroeconomics  magnitude  map-territory  map  market_research  math.ds  math  mathematics  maxim-gun  mechanics  media:document  medicine  mediterranean  mental-math  messages  meta:medicine  meta:prediction  meta:rhetoric  meta:science  metabolic  metabuch  metaheuristics  metameta  metaphysics  methodology  metrics  michael_mcderment  military  minimalism  model-organism  model-selection  modeling  models-and-modes  models  mongodb  morality  mostly-modern  motivation  multi  multiobjective-optimization  music  mutation  nature  netcultures  network-theory  networks  neural-networks  neuro  new-religion  news  nibble  nietzschean  nihil  nitty-gritty  no-go  noise-structure  nonlinearity  novelty  nudge-targets  numerical-methods  occam's-double-edged-razor  occam's_razor  occam  ockhams-razor  omit-needless-words  ontario  oop  opensource  operational  operations-research  order-disorder  orfe  org:anglo  org:bleg  org:edu  org:gov  org:junk  org:mag  org:nat  org:rec  org:sci  org:theos  oscillation  outcome-risk  overflow  parasites-microbiome  pareto-gp  party-like-it's-1995  patterns  pdf  people  performance-measure  performance  pharma  philosophy-of-science  philosophy  philosophy_of_science  phm  phylogenetics  physics  piracy  pivots  play  poast  policy  polisci  politics  population-genetics  possumpalace  postmortem  postrat  pragmatism  pre-2013  prepping  pro-rata  probability  productivity  profile  protestant-catholic  prudence  psychology  python  q-n-a  qra  qtl  quality  questions  queue  quote  quotes  r  radford-neal  radford_neal  random  rant  rat-pack  rather-interesting  rationality  ratty  reason  red-queen  reduction  reference  reflection  regularizer  religion  research-article  research  rest  review  rhetoric  risk  robust  rock  roots  rot  ruby  rules  russia  s:**  s:*  sapiens  scalability  scale  scaling-up  science  scitariat  search  selection  semanticweb  sequence-analysis  server  sex  sexuality  shell  shift  signal-noise  signaling  simplicity  skeleton  sky  slides  small_business  smoothness  soap  social-science  social  sociology  soft-question  space  sparql  spatial  spearhead  speculation  speed  spreading  standards  stat-mech  statesmen  statistics  stats  stochastic-resonance  stories  strategies  strategy  street-fighting  structural-biology  structure  study  style  stylized-facts  summary  survey  syntax  synthesis  systematic-ad-hoc  talks  taxonomy  teaching  technology  techtariat  testing  tetlock  that-gödel-fellow-said-something-relevant  the-bones  the-classics  the-great-west-whale  the-trenches  theory-practice  theory  theos  thermo  thick-thin  thinking  threat-modeling  tidbits  time-series  time  tinderboxed  traces  trends  trivia  troll  truth  twitter  ui  unaffiliated  uncertainty  unintended-consequences  unix  usa  vague  variance-components  visual-understanding  visualization  vocabulary  vulgar  walter-scheidel  war  waves  web  webdev  webservices  west-hunter  white-paper  wiki  wire-guided  words  wordsmiths  world-war  world  writing  wsgi  yvain  zope  🌞  🎩  🤖  🦀 

Copy this bookmark: