3 ms·

> With big data you don't need models, you have actual real time information about the entire system. Your comment makes absolutely no sense. Models are just

• x5n1 10y ago
  > Models are just means to predict the future based on observations made in the present and/or in the past. Models are representations of reality. If you have the actual reality, as in all transactions for every product made by every agent, you don't need a model as you have the actual data. You can watch that in real time, and you can also test hypothesis and theories in real time. That in opposition to a model, say an economic model, which would be based on some theoretical assumptions attempting to predict supply and demand while looking at certain variables. If you have all the transactions that every single agent makes, as well as all social-economic variables about that person in real time, then any theory you have about their behavior can be tested in real time, rather than against a model, and validated or thrown away. > A model is a tool that tells you how systems evolve and lets you infer what you can expect out of them in order to make informed decisions. Models are representations of reality, rather than simple inference it's much more powerful to have actual data from reality available to test your theories, not just a model. Models are simplified versions of reality, a reality which is forever changing. Economic models are often wrong. > Models are used to make decisions based on how things are expected to be years and decades into the future. It makes absolutely no sense to claim that you can tell how things will be in the immediate future if you're trying to make decisions based on what will happen over the following decades. And this is why economics is often wrong. Knowing what is happening right now and responding to it, is much more powerful than making theoretical assumptions about how things are and then waiting and praying that they are right... which often they are not. > The free market is just a specific theoretical definition of an economic system. It establishes conditions, draws relationships between factors and their effects, and provides a tool to model the evolution of economies. I don't know which tools you are talking about. I don't know what evolution you are talking about. One reason the free market is superior to central planners is exactly because central planners did not have actual market data but instead had to rely on models for supply and demand. When central planners have access to real time actual market demand as well as all data about each agent, it makes it much more likely that they will succeed at their job than the way the Soviet Union did it. Also your tone stinks, work on that.
  • bubuga 10y ago
    You clearly have no clue about what you're talking about. You obviously don't understand what a model is, let alone how and why it's used, and you clearly don't understand the role models play in decision-making processes. Furthermore, you are completely oblivious to the reality and application of models in general and economic models in particular. If you are entirely oblivious to the issue you're trying to discuss, it's pointless to continue discussing it with you.
    • x5n1 10y ago
      See you have this model: https://en.wikipedia.org/wiki/Computable_general_equilibrium https://en.wikipedia.org/wiki/Computable_general_equilibrium See this model "estimates how an economy might react to changes in policy, technology or other external factors." This model uses real data. Why use this model at all? Why not change the policy itself and see the effects of it in real time? If you can do that as you can in terms of some decisions which are reversible. If on the other hand that's not possible, why not validate this model continuously against the real market with every single policy to point to its weaknesses and failures? That's what big data allows you to do. That's all I was getting at.
      • donkalypse 10y ago
        Agile fiscal and monetary policy? Seems dangerous.