Can Mathematical models predict the economy?



14102013, 11:40 PM




RE: Can Mathematical models predict the economy?
(17092013 04:27 AM)Bucky Ball Wrote: The influence of all possible variables which could impact an economic decision are accounted for, mathematically. Yes, ideally but that never actually happens in practice. Economic modellers endeavour to identify all relevant variables but they never do so and one model is displaced by another, richer, model. This is well illustrated in the work on asset pricing models which won the Nobel prize for Eugene Fama, Lars Peter Hansen and Robert Shiller which you linked to in another post. Fama's ThreeFactor Model (TFM) is an improvement on the Capital Asset Pricing Model (CAPM) which won William Sharpe the Nobel prize in 1990. Just because the CAPM won Sharpe the prize doesn't mean that "all possible variables which could impact an economic decision are accounted for" in the CAPM. They weren't. The CAPM described asset return only in terms of a single variable and the entire equation is composed of four variables.[1] The TFM gets its name from the fact that it describes asset return in terms of three variables instead of one; also it is composed of ten variables.[2] The TFM can can only be regarded as provisional, the best model of asset returns that we have now. There is no reason to believe that the TFM is complete and that it takes account of all relevant variables. The history of econometric models is one of incrementally closer approximations of reality and there are no econometric models that describe the realworld perfectly. Your claim that "The influence of all possible variables which could impact an economic decision are accounted for, mathematically" is made from a position of ignorance and arrogance. There are no such econometric models. If you disagree then present them to us. There are philosophical debates about the possible limits of qualitative economic as well as econometric models (see for example Philosophy of Economics and BryanS and JAH have mentioned some specific issues but you just dismissed them as "idiots"). Quote:You have no clue what you are talking about. But thanks for proving yet again, what am imbecile you are. I don't think you have any idea about what you are talking about. Quote:You've obviously never even seen an econometric model. That appears to be the extent of your knowledge of econometrics, i.e. that you have merely seen at least a set of equations. 

15102013, 05:42 PM




Can Mathematical models predict the economy?
The economy and stock market are fundamentally unpredictable.
First of, the economy depends on many variables that lie outside of the economy. In order to predict the economy, one has to predict the weather a long time in advance, foresee the assassinations of heads of state, etc. also, as soon as a model is found that predicts the economy and the stock market, some people will implement it in a computer and use massive leverage to get monetary gains from it. By doing so, they change the dynamics of the economy itself, which will invalidate the model. Because powerful computers are doing the big part of the trading, the stock market is probably less predictable than ever. This became obvious when the dow jones index dropped by thousand points or so because of one computer glitch. 



« Next Oldest  Next Newest »

User(s) browsing this thread: 1 Guest(s)