simardcasanova’s avatarsimardcasanova’s Twitter Archive—№ 26,217

                1. …in reply to @colette_salemi
                  @colette_salemi There is a probably a couple of explanations I’d argue that the first one is that the first generation of ABMs were developed to explicitly “kill” mainstream econ. Hence their lukewarm reception…
                  Permalink simardcasanova’s avatar ❤️ 11 Favorites Mood 0
              1. …in reply to @simardcasanova
                @colette_salemi Another issue with ABMs is that most of the time, they don’t have closed analytical solutions (and when they do, it’s only on a simplified version of the model, as emphasized in this paper sciencedirect.com/science/article/abs/pii/S0167268119300228) It’s not a bad thing per see, but economists don’t like that
                Permalink simardcasanova’s avatar ❤️ 9 Favorites Mood 0
            1. …in reply to @simardcasanova
              @colette_salemi The question of their empirical validation is another open issue, still not resolved yet (albeit progresses are being made) And I would also argue that they lack reproducibility and are often (rightfully imo) seen as black boxes
              Permalink simardcasanova’s avatar ❤️ 4 Favorites Mood -2 🙁
          1. …in reply to @simardcasanova
            @colette_salemi Last but not least, the natural application domain for ABMs is macro. And macro is subject to a ton of power struggles, not related to research itself but rather to academic prestige. I see this as the main reason why funding for ABMs has been so sparse
            Permalink simardcasanova’s avatar ❤️ 4 Favorites Mood -1 🙁
        1. …in reply to @simardcasanova
          @colette_salemi And without funding, the methodology is only slow to improve That being said, the Bank of England is funding more ABM research, so I’m expecting the methodology to improve more rapidly
          Permalink simardcasanova’s avatar ❤️ 5 Favorites Mood +4 🙂
      1. …in reply to @simardcasanova
        @colette_salemi Younger generations of ABM economists (me included) are also less keen to “kill” mainstream econ I guess we have less extreme discourse about ABMs and hence it’s probably easier to attract more researchers into the methodology
        Permalink simardcasanova’s avatar ❤️ 3 Favorites Mood +2 🙂
    1. …in reply to @simardcasanova
      @colette_salemi Those are just my 50 cents, I’m not pretending I have all the answers! But I’m convinced ABMs are in a good position to radically improve a ton of econ literature in the upcoming decade or so. As you said in your first tweet, they can really augment equation-based models!
      Permalink simardcasanova’s avatar ❤️ 3 Favorites Mood +7 🙂
  1. …in reply to @simardcasanova
    @colette_salemi (Especially because heterogenous agents are so easy to models, and also because equilibrium is not an “assumption” but a result, and thus they can tolerate much more easily non-equilibrium dynamics than regular equation-based models)
    simardcasanova’s avatar ❤️ 3 Favorites Mood +1 🙂