Thread: Derive equation from stat plot

I'm kind of lost on how to go about this but:

We were given a project and we have to derive our own equations from the data we were given.

I have a graph that is a parabolic shape and without using the stat>calc option I have to derive an equation, suggestions? I'd rather not give all the data and ask someone to solve, just confused on how to begin deriving it.

If you need anymore information please ask, for the record we haven't gotten past limits so I'm not sure if we're suppose to use derivatives or not.

Originally Posted by doomgaze

I'm kind of lost on how to go about this but:

We were given a project and we have to derive our own equations from the data we were given.

I have a graph that is a parabolic shape and without using the stat>calc option I have to derive an equation, suggestions? I'd rather not give all the data and ask someone to solve, just confused on how to begin deriving it.

If you need anymore information please ask, for the record we haven't gotten past limits so I'm not sure if we're suppose to use derivatives or not.

The TI graphing calculator should have the built-in feature that finds an equation model to fit your data. Populate your list 1 and list 2, then look under the stats option. It's in the same place as Linear Regression... might be PowerReg.

Good luck!

well she mentions that the equations should be derived "by hand", so I assuming that's not allowed.

okay. then let the quadratic be f(x) = ax^2 + bx + c

Pick any three points from your data, and plug them in one at a time. For each point you will get a condition involving a, b, and c.

Set up the three conditions as a system of linear equations and solve for a, b, and c using elimination/substitution/matrices. Once you have the coefficients, plot the graph on top of the scatter plot to make sure it describes the data well.

I hope this helps!