Exponential Regressions

I was wondering if it is possible in StatistiXL to calculate SE of "a", SE of "b", and the SE of the estimate for an exponential regression of the form y = a.e^bx. I would be grateful if anyone who has knowledge about this can provide some information...

Cheers, Tim

Comments

  • Hi

    statistiXL currently only supports Linear Regression, not Exponential Regression. The only option open to you would be to transform your data to its linear equivalent. In this case I think you would need to take natural logs of both sides

    Ln(Y) = ln(a) + ln(e^bX) =>
    Ln(Y) = ln(a) + bX

    So, plot ln(Y) against ln(X) and you get ln(a) and b.

    Note however that while this gives the SE of b you will get the SE of ln(a)! And of course the SE of the estimate is for the Ln-transformed equation, NOT the raw equation.

    Best Regards

    Alan
Sign In or Register to comment.