Thanks for the provided info! If I understand it correctly, then in order to apply the least square method for regression analysis the data should be mapped to a linear model in order to get better results? Converting the data to ln y indeed gives better results, but is still not a perfect fit like I have for instance with polynomial functions. Are there other methods available for curve fitting which don't have this requirement?
regards, Christiaan -- View this message in context: http://apache-commons.680414.n4.nabble.com/Math-Exponential-curve-fitting-problem-tp2995629p2996928.html Sent from the Commons - User mailing list archive at Nabble.com. --------------------------------------------------------------------- To unsubscribe, e-mail: user-unsubscr...@commons.apache.org For additional commands, e-mail: user-h...@commons.apache.org