1

I've obtained a trend line from a particular set of data. What I'd like to do now is to reuse this trend line to predict values from a given pair (x,y) of coordinates.

To put it another way, I have one pair (x,y) that I know is correct for sure. I don't know any other point. Let's assume the behavior of this new set is similar to the one I've got the trend line from. Is there any way Excel could compute other points following this trend line?

0

First, show the equation for the trendline, via Format Trendline->Display Equation on Chart

Then convert the shown equation to a formula, so that it will produce a Y for any given X.

Note though, your original data is probably on both sides of the trendline (i.e. the trendline is an estimated function), your new data will all be exactly on the new trendline (as it's actually generated from a known function).

  • Well, you're right. Actually my approach was wrong. The trend line is not of any help here. – milko May 28 '14 at 18:34
  • What are you really trying to do? do you need to fake up realistic looking data that ends up with the same trendline as another set of data (while still looking random)? – Madball73 May 28 '14 at 19:37
  • I have collected some points from "measures". What I noticed is if x double, the corresponding y value is tripled. It's not actually true but close enough for a first approximation. However what I'd like to do is to get the "real" pattern corresponding to the trend of my data. So with the other (x,y) pair I've got (x and y are both known) I would be able to get different values but following the same "rule". If we consider a simple example, let's say (x1,y1) = (2,3) then (x2,y2) = (4,9). Now let's say (x1,y1) = (2,5) with the same rule we get (x2,y2) = (4,15). My goal is to find this "rule" – milko May 28 '14 at 20:10

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.