Representation of regression in a matrix form: $$Y = XA + E,$$

where: $X$ - independent variables, $Y$ - dependent variables (observations), $E$ - errors, which have a uniform distribution, $A$ - regression parameter.

How to evaluate the parameters $A$ of this multivariate regression using the maximum likelihood method with a uniform distribution of errors?

  • $\begingroup$ Is the range of the uniform distribution given or does it need to be estimated? What do you assume about the independence (or lack thereof) of $E$? $\endgroup$ – whuber May 27 '19 at 11:45
  • 1
    $\begingroup$ @whuber Thank you for the answer! Given: E = [e1, e2, ..., en] ’, // vector of residuals Where ei = U [-ui, ui]; The residuals ei have a uniform distribution with a known range [-ui, ui], All residuals are independent. I need to estimate the vector of parameters A and its variance. $\endgroup$ – Minh Đại May 27 '19 at 16:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.