Skip to content
By using Twitter’s services you agree to our Cookies Use. We and our partners operate globally and use cookies, including for analytics, personalisation, and ads.

This is the legacy version of twitter.com. We will be shutting it down on June 1, 2020. Please switch to a supported browser, or disable the extension which masks your browser. You can see a list of supported browsers in our Help Center.

  • Home Home Home, current page.
  • About

Saved searches

  • Remove
  • In this conversation
    Verified accountProtected Tweets @
Suggested users
  • Verified accountProtected Tweets @
  • Verified accountProtected Tweets @
  • Language: English
    • Bahasa Indonesia
    • Bahasa Melayu
    • Català
    • Čeština
    • Dansk
    • Deutsch
    • English UK
    • Español
    • Filipino
    • Français
    • Hrvatski
    • Italiano
    • Magyar
    • Nederlands
    • Norsk
    • Polski
    • Português
    • Română
    • Slovenčina
    • Suomi
    • Svenska
    • Tiếng Việt
    • Türkçe
    • Ελληνικά
    • Български език
    • Русский
    • Српски
    • Українська мова
    • עִבְרִית
    • العربية
    • فارسی
    • मराठी
    • हिन्दी
    • বাংলা
    • ગુજરાતી
    • தமிழ்
    • ಕನ್ನಡ
    • ภาษาไทย
    • 한국어
    • 日本語
    • 简体中文
    • 繁體中文
  • Have an account? Log in
    Have an account?
    · Forgot password?

    New to Twitter?
    Sign up
NYarvin's profile
Norman Yarvin
Norman Yarvin
Norman Yarvin
@NYarvin

Tweets

Norman Yarvin

@NYarvin

yarchive.net/blog
Joined April 2013

Tweets

  • © 2020 Twitter
  • About
  • Help Center
  • Terms
  • Privacy policy
  • Imprint
  • Cookies
  • Ads info
Dismiss
Previous
Next

Go to a person's profile

Saved searches

  • Remove
  • In this conversation
    Verified accountProtected Tweets @
Suggested users
  • Verified accountProtected Tweets @
  • Verified accountProtected Tweets @

Promote this Tweet

Block

  • Tweet with a location

    You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more

    Your lists

    Create a new list


    Under 100 characters, optional

    Privacy

    Copy link to Tweet

    Embed this Tweet

    Embed this Video

    Add this Tweet to your website by copying the code below. Learn more

    Add this video to your website by copying the code below. Learn more

    Hmm, there was a problem reaching the server.

    By embedding Twitter content in your website or app, you are agreeing to the Twitter Developer Agreement and Developer Policy.

    Preview

    Why you're seeing this ad

    Log in to Twitter

    · Forgot password?
    Don't have an account? Sign up »

    Sign up for Twitter

    Not on Twitter? Sign up, tune into the things you care about, and get updates as they happen.

    Sign up
    Have an account? Log in »

    Two-way (sending and receiving) short codes:

    Country Code For customers of
    United States 40404 (any)
    Canada 21212 (any)
    United Kingdom 86444 Vodafone, Orange, 3, O2
    Brazil 40404 Nextel, TIM
    Haiti 40404 Digicel, Voila
    Ireland 51210 Vodafone, O2
    India 53000 Bharti Airtel, Videocon, Reliance
    Indonesia 89887 AXIS, 3, Telkomsel, Indosat, XL Axiata
    Italy 4880804 Wind
    3424486444 Vodafone
    » See SMS short codes for other countries

    Confirmation

     

    Welcome home!

    This timeline is where you’ll spend most of your time, getting instant updates about what matters to you.

    Tweets not working for you?

    Hover over the profile pic and click the Following button to unfollow any account.

    Say a lot with a little

    When you see a Tweet you love, tap the heart — it lets the person who wrote it know you shared the love.

    Spread the word

    The fastest way to share someone else’s Tweet with your followers is with a Retweet. Tap the icon to send it instantly.

    Join the conversation

    Add your thoughts about any Tweet with a Reply. Find a topic you’re passionate about, and jump right in.

    Learn the latest

    Get instant insight into what people are talking about now.

    Get more of what you love

    Follow more accounts to get instant updates about topics you care about.

    Find what's happening

    See the latest conversations about any topic instantly.

    Never miss a Moment

    Catch up instantly on the best stories happening as they unfold.

    1. stucchio‏ @stucchio May 26
      • Report Tweet
      • Report NetzDG Violation

      Interesting paper that discusses breaks down a *cultural* distinction in stats: gambler vs scientist. Most Bayesians are gamblers and Frequentists are scientists, but nothing fundamental to this. Just cultural. https://buff.ly/2Zx8CEU  ht @georgizgeorgiev

      4 replies 0 retweets 5 likes
    2. Georgi Z. Georgiev‏ @georgizgeorgiev May 26
      • Report Tweet
      • Report NetzDG Violation
      Replying to @stucchio

      Everyone is a speculator by necessity. Frequentists (or I at least) want data on one side, decision-making and therefore risk/reward analysis on the other, separate. This way anyone can impose any loss function they like on the data and its uncertainty estimate(s).

      1 reply 0 retweets 0 likes
    3. Georgi Z. Georgiev‏ @georgizgeorgiev May 26
      • Report Tweet
      • Report NetzDG Violation
      Replying to @georgizgeorgiev @stucchio

      I can understand why one would be tempted to assume all the risk and reward data you need is in the test data itself (thus trivial loss functions may apply). However this is true for a very, very limited set of application domains (a prominent one, I think, is stock trading).

      1 reply 0 retweets 0 likes
    4. stucchio‏ @stucchio May 27
      • Report Tweet
      • Report NetzDG Violation
      Replying to @georgizgeorgiev

      On the mathematical level Bayesians do the same: opinions about the world (the posterior) on one side, decisions minimizing expected loss on the other. And you can let the loss function be as complicated as you want: in the general case the posterior is calculated with MCMC. 1/

      1 reply 0 retweets 0 likes
    5. stucchio‏ @stucchio May 27
      • Report Tweet
      • Report NetzDG Violation
      Replying to @stucchio @georgizgeorgiev

      On the flip side, Frequentists *can* do the same via minimax with an equally complex loss function. They might even get a computational advantage if they can draw samples from the likelihood with quasimontecarlo rather than MCMC. But it's not that common. 2/

      1 reply 0 retweets 0 likes
    6. stucchio‏ @stucchio May 27
      • Report Tweet
      • Report NetzDG Violation
      Replying to @stucchio @georgizgeorgiev

      And on the flip side, Bayesian techniques can provide guarantees quite similar to frequentist ones. Here's an example: https://www.dropbox.com/s/w7x8ufvgrul5grm/Frequentist_accuracy_of_Bayesian_estimates__rssb12080.pdf?dl=0 … But this is also not that common. Concentration bounds are used to bound loss, not compute frequentist guarantees. 3/

      1 reply 0 retweets 0 likes
    7. stucchio‏ @stucchio May 27
      • Report Tweet
      • Report NetzDG Violation
      Replying to @stucchio @georgizgeorgiev

      And this is why I'm speculating that a big chunk of the disagreement among practitioners is more cultural than mathematical. Specifically, there's one school (B) that tries to make money by gambling and a different school (F) that tries to avoid making mistakes. 4/

      1 reply 0 retweets 0 likes
    8. stucchio‏ @stucchio May 27
      • Report Tweet
      • Report NetzDG Violation
      Replying to @stucchio @georgizgeorgiev

      The mathematical techniques are almost an afterthought. When I use frequentist techniques I immediately jump to decision theory, minimax, etc. I get the impression you and many others don't. That's why I'm speculating that Bayes vs Freq is actually about something else. 5/5

      1 reply 0 retweets 0 likes
      Norman Yarvin‏ @NYarvin May 27
      • Report Tweet
      • Report NetzDG Violation
      Replying to @stucchio @georgizgeorgiev

      To me, the difference is one of social position. If you're trying to humbly beseech the powers that be, you'll be saying "But this is what actually happens in practice" (frequentist). If you're in charge, you aren't shy about using your prior knowledge (Bayesian).

      5:01 PM - 27 May 2020
      1 reply 0 retweets 0 likes
        1. New conversation
        2. stucchio‏ @stucchio May 27
          • Report Tweet
          • Report NetzDG Violation
          Replying to @NYarvin @georgizgeorgiev

          I don't think this is right. Consider the classic xkcd regarding the Bayes vs Frequentist answer to "did the sun explode?" https://xkcd.com/1132/  The comic is a straw man of frequentist stats. But Frequentists handle cases like this by incorporating prior knowledge.pic.twitter.com/ubVByapM2o

          2 replies 0 retweets 0 likes
        3. Norman Yarvin‏ @NYarvin May 27
          • Report Tweet
          • Report NetzDG Violation
          Replying to @stucchio @georgizgeorgiev

          That "frequentist" is just wrong by any standard. It's not that "the probability of this result happening by chance is X", it's "if the result were chosen by pure chance, the probability of this result happening would be X".

          2 replies 0 retweets 0 likes
        4. 3 more replies

      Loading seems to be taking a while.

      Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

        Promoted Tweet

        false

        • © 2020 Twitter
        • About
        • Help Center
        • Terms
        • Privacy policy
        • Imprint
        • Cookies
        • Ads info