Error in polynomial trend line equation

  1. The Order 5 Polynomial (Quintic) and Order 4 Polynomial (Quartic) don't appear on the trend line equation. See comparison with Microsoft Excel formula - which is correct.
  2. The Order 5 Polynomial equation is incorrect.

Posted on Jun 29, 2024 12:38 AM

Reply
Question marked as Top-ranking reply

Posted on Jun 29, 2024 4:11 AM

I tried to recreate your chart and I can verify what you are saying about the trendline equation not showing the higher powers. It isn't that way with all data/charts but it is with this data. File a bug report.

Numbers->Provide Numbers Feedback.


Can you provide the actual data used to create this chart and trendline. I cannot verify differences in the equations between Excel and Numbers without having the actual data you used. When I pick points by eye from the trendline to use as the "data", the resulting equation is far different.


Aside from these problems, you might consider that this trendline is probably a poor representation of the data except on this short interval. It probably goes shooting off in some wild directions outside of this interval. Also, the equation shown in Excel is not all that accurate. If you try to use it to create data points, you will get wildly inaccurate results. One clue is that the exponents on the higher powers are exactly 1E-24, 2E-18, and so on which is highly unlikely. Try creating some data from that trendline equation and see what you get. I tried it with my eyeballed data and it is totally useless. If you provide the actual data, I might try calculating the actual exponents using LINEST just to see what they really are.


2 replies
Question marked as Top-ranking reply

Jun 29, 2024 4:11 AM in response to Antipodean1788

I tried to recreate your chart and I can verify what you are saying about the trendline equation not showing the higher powers. It isn't that way with all data/charts but it is with this data. File a bug report.

Numbers->Provide Numbers Feedback.


Can you provide the actual data used to create this chart and trendline. I cannot verify differences in the equations between Excel and Numbers without having the actual data you used. When I pick points by eye from the trendline to use as the "data", the resulting equation is far different.


Aside from these problems, you might consider that this trendline is probably a poor representation of the data except on this short interval. It probably goes shooting off in some wild directions outside of this interval. Also, the equation shown in Excel is not all that accurate. If you try to use it to create data points, you will get wildly inaccurate results. One clue is that the exponents on the higher powers are exactly 1E-24, 2E-18, and so on which is highly unlikely. Try creating some data from that trendline equation and see what you get. I tried it with my eyeballed data and it is totally useless. If you provide the actual data, I might try calculating the actual exponents using LINEST just to see what they really are.


This thread has been closed by the system or the community team. You may vote for any posts you find helpful, or search the Community for additional answers.

Error in polynomial trend line equation

Welcome to Apple Support Community
A forum where Apple customers help each other with their products. Get started with your Apple Account.