Re: Econometrics and Machine Learning

I am not quite sure whether I understand you correctly but if you mean
learning algorithms in neural networks (NN), my opinion is this:
NN might fit your data much better, but that doesn't imply that the
predictive/forecasting properties are any better. Linear regression
models allow are straighforward and imply a structural form. nonlinear
regressions are even better and closely related to NN.

Dirk

.

Relevant Pages

  • Re: Logistic Regression
    ... I did manage to fit some ... logistic model to the data because of the large number of dimensions. ... least squares linear regression first, and see how that works out on ...
    (sci.stat.math)
  • Re: Origin of Chinese spoken languages
    ... You said to apply a simple linear regression, and I did (as well as ... simple exponential growth model, and it gave me a better fit (what do ... the number of characters change everytime? ...
    (sci.lang)
  • Re: Origin of Chinese spoken languages
    ... > solution and I suggested a simple linear regression. ... Why don't you look at your data then, and provide a suitable fit for the ... is it more appropriate to compare residuals on an absolute versus ... growth for chinese characters occurred around the Warring states period ...
    (sci.lang)
  • Re: Origin of Chinese spoken languages
    ... > solution and I suggested a simple linear regression. ... So I try to fit to ... Shuowen Jiezi has 9353 characters. ... > errors (residuals). ...
    (sci.lang)
  • Re: Comparing fits
    ... One measure of goodness of fit for linear regression is to take the ... correlation coefficient between y and x1, ... variable that gives the smallest sum of squared errors. ...
    (sci.stat.math)