econometrics   2062

« earlier    

Algorithmic Fairness
Concerns that algorithms may discriminate against certain groups have led to numerous efforts to 'blind' the algorithm to race. We argue that this intuitive perspective is misleading and may do harm. Our primary result is exceedingly simple, yet often overlooked. A preference for fairness should not change the choice of estimator. Equity preferences can change how the estimated prediction function is used (e.g., different threshold for different groups) but the function itself should not change. We show in an empirical example for college admissions that the inclusion of variables such as race can increase both equity and efficiency.
econometrics  algorithms  ethics  machine_learning  sendhil.mullainathan 
3 days ago by rvenkat
This is exactly what I also claim in the Identification Zoo. Except for try DAGs part.

Causal_Inference  econometrics 
4 weeks ago by demetriodor
[1803.06386] Forecasting Economics and Financial Time Series: ARIMA vs. LSTM
"Forecasting time series data is an important subject in economics, business, and finance. Traditionally, there are several techniques to effectively forecast the next lag of time series data such as univariate Autoregressive (AR), univariate Moving Average (MA), Simple Exponential Smoothing (SES), and more notably Autoregressive Integrated Moving Average (ARIMA) with its many variations. In particular, ARIMA model has demonstrated its outperformance in precision and accuracy of predicting the next lags of time series. With the recent advancement in computational power of computers and more importantly developing more advanced machine learning algorithms and approaches such as deep learning, new algorithms are developed to forecast time series data. The research question investigated in this article is that whether and how the newly developed deep learning-based algorithms for forecasting time series data, such as "Long Short-Term Memory (LSTM)", are superior to the traditional algorithms. The empirical studies conducted and reported in this article show that deep learning-based algorithms such as LSTM outperform traditional-based algorithms such as ARIMA model. More specifically, the average reduction in error rates obtained by LSTM is between 84 - 87 percent when compared to ARIMA indicating the superiority of LSTM to ARIMA. Furthermore, it was noticed that the number of training times, known as "epoch" in deep learning, has no effect on the performance of the trained forecast model and it exhibits a truly random behavior."
econometrics  time-series  lstm  arima 
9 weeks ago by arsyed

« earlier    

related tags

#reading  ?  2015  80000-hours  _classroom_  _rm_  administrative_state  advertising  advice  africa  age-generation  algorithms  analysis  antidemos  applied-economics  archaeology  arima  arrows  article  articles  asia  athey  atmosphere  authoritarianism  behavior  behavioral-econ  best-practices  big_data  biophysical-econ  blogs  book  books  branches  broad-econ  career  caste_system  causal-inference  causal_inference  causality  causation  china  cities  civic  class  climate-change  coalitions  cocktail  code  cog-psych  cohesion  college  commentary  communism  comparative  comparison  compensation  confounding  contemporary_culture  control  convexity-curvature  correlation  corruption  cost-benefit  counterfactual  course  crime  critique  cryptocurrency  culture  current-events  curvature  data  database  death  demographics  developing-world  development_economics  developmental  diversity  dmce  draw  dropbox  econ-metrics  econ-productivity  econometric-analysis  economic_geography  economic_history  economic_sociology  economics  education  effect-size  elections  elite  empirical  empirics  endo-exo  endogenous-exogenous  entropy-like  environment  ethanol  ethics  events  evidence-based  example  examples  excel  experimental_sociology  experiments  explanation  feudal  finance  financial-econometrics  fitness  fitsci  fluid  for_friends  foreign-policy  free  gdp  gender  geography  governance  government  grad-school  graph  gretl  group-level  growth-econ  hanushek  hari-seldon  have_read  health  heckman-sample-selection  higher-ed  history  hive-mind  homophily  human-capital  human  hypochondria  hypothesis-testing  i_remain_skeptical  ideas  ideology  imbens  immigration  incentives  income  india  industrial-org  inequality  infoviz  innovation  input-output  institutions  instrumental-variable  instrumental-variables  instrumental_variables  interaction  interactions  interactive  interesting  interests  intervention  intricacy  iq  iv  julia  korea  labor  latin-america  lead-crime_hypothesis  leadership  learning  lectures  leviathan  life-history  list  loglog  long-term  longitudinal  lstm  m&s  machine-learning  machine  machine_learning  machinelearning  macro  management  map-territory  math  matlab  measurement  meta-analysis  methodology  methods  micro  microeconomics  microfoundations  migration  min_wage  mobility  model-class  model  modeling  modelling  models  moderation  moment_inequalities  money  mooc  mostly-modern  multi  nationalism-globalism  natural-experiment  natural_experiment  networks  news  nihil  nitty-gritty  nonlinearity  north-weingast-like  northeast  notes  null-result  oil-markets  ordered-logistic-regression  org:local  org:mag  org:ngo  org:rec  organizing  pandas  panel-data  paper  papers  parenting  path-analysis  path  paying-rent  pdf  peace-violence  people  phd  piracy  planning  platform_economics  polarization  policing  policy-evaluation  policy  policy_as_a_social_process  polis  polisci  political-econ  political_economy  political_science  politics  pop-diff  population  portable  portableapps  poverty  preprint  presentation  propensity-matching-score  property-rights  proposal  psychology  public_administration  public_goods  public_policy  python  qed  quantitative  r-project  r  race  racism  raj.chetty  ranking  ratty  re:adafaepov  realpolitik  regression-discontinuity-design  regression  regret  rent-seeking  replication-files  research  research_methods  review  rhythm  rindermann-thompson  rmii  roots  rot  rstats  safety  scale  schools  science  selection  sendhil.mullainathan  series  shift  signaling  simultaneous-equations-model  sinosphere  sky  slides  social-capital  social-choice  social-norms  social-structure  social  social_construction_of_knowledge  socioeconomic  sociology  software  spatial_statistics  spatio-temporal_statistics  spearhead  speculation  sports  stata  statesmen  statistics  stats  study  stylized-facts  sub-sample  subgroup-analysis  subgroup  suest  summary  suppliers  supply-demand  survey  susan.athey  susanathey  symposium  talks  taxes  teaching  technology  texas  textbook  the-south  the-world-is-just-atoms  time-series-econometrics  time-series  time-use  time  tips  to:nb  to_teach:undergrad-ada  top-n  totwitter  trade  tradeoffs  trends  tricks  trivia  trust  twitter  unintended-consequences  unions  united_states_of_america  usa  violence  wapo  wealth-of-nations  wealth  welfare  white-paper  wikipedia  wonkish  wooldridge  world  world_trends  🌞  🎓  🎩 

Copy this bookmark: