I estimate ARIMA models or another models that explain now price with past pirce. If I use this models , already ı reject efficient market hypothesis? I write master thesis and my teacher asked me what is your hypothesis.Can I say my empricial results are succesful ,the predicting performance is good and accetable so the market is not efficient


closed as unclear what you're asking by whuber Aug 13 at 20:06

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • 2
    $\begingroup$ What is the result that makes you reject the EMH ? The only real way to reject it would be to show that you can profit, after transaction costs, by using ARIMA models to predict the price ( or the return ). If you've done that, then it's best not to publish the thesis and trade off of it !!!!! $\endgroup$ – mlofton Aug 10 at 8:30
  • $\begingroup$ My models' performans criterias (MSE,RMSE,MAPE,r2) is good. So the result is evidence of predictable. So if we can predict the market then the market is not efficient. Or ıf ı write opposite this stuaiton , musnt be my assumption that market is not efficient. Because if I accept it is efficient then it is not logicial to try predicting, isnt it? Thank you for your answer $\endgroup$ – murat tuna Aug 10 at 8:58
  • 1
    $\begingroup$ Change your outcome variable from the price $y_t$ to the return $\log y_t - \log y_{t-1}$ and try and predict that with any variable available at $t-1$ and earlier. What MSE do you get for that? The EMH is about return predictability $\endgroup$ – CloseToC Aug 10 at 9:26
  • $\begingroup$ If you predict prices then already dont you find predict? $\endgroup$ – murat tuna Aug 10 at 9:42
  • 1
    $\begingroup$ @murat tuna: You can't use the MSE or any other loss criteria like that to accept or reject EMH. You have to create a portfolio ( long when prediction is positive and short when prediction is negative ) based on your predictions and then check if it's profitable after transaction costs are taken into account. If you can do this, then you have shown that the EMH should be rejected. People have been trying to do this for years and years and few are successful. Also, CloseToC is correct that your'e way better off trying to predict returns than trying to predict price. $\endgroup$ – mlofton Aug 11 at 18:05