Quadratic Trendline Equation is Completely wrong, or did I interpret it wrongly?
Hello,
I have plotted some data (shown in file "2A2"), and tried to fit lines to it, getting an okay fit using quadratic. But I realise it is completely wrong when I did a plot of the equation (in file "plot"). y = -0.007x2 + 7333.9x - 2E+09 is a "frown", and has turning point at x = 523850, meaning a negative gradient for all x > 523850. But my data is not this way!
Thanks for your help!!
Hi,
1) you got just 1 meaningful digit om x^2 and x^0 partst so it could be for instance y = -0.00749999x^2 + 7333.9x - 2.49999E+09 or may be y = -0.00650000x^22 + 7333.9x - 1.5000000E+09 etc.
2) your data starts from some 524169.4 and ends at some 524180.8 so may be drawnimg from 524165 to 524185 would be reasonable, but extending fitted equation from ca 160000 to ca 970000 is for sure simply ABUSE.
ad 1: to get better (more significant digits) assesment of fitted equation see for instance: https://www.excelforum.com/excel-gen...fficients.html
Best Regards,
Kaper
Re: Quadratic Trendline Equation is Completely wrong, or did I interpret it wrongly?
Thank you Kaper. I didn't believe it but after extending my number of decimal places to 30, I can finally see the curve in its entirety (I just extend the domain just to see/ demonstrate the curve shape). Thanks a lot!
Glad to hear that.
If that takes care of your original question, please select Thread Tools from the menu link above and mark this thread as SOLVED. Thanks.
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
LinkBack URL
About LinkBacks
Register To Reply
Bookmarks