Comparing Two curves (two sets of data)

[ptc-1441311]

I would like to know if there is a method in stats where we can compare the two curves (data set A & data set B) with different number of points?

I would like to know if there is a significant difference between the data I obtained using a model equation (with number of points = 150) to the experimental data (with number of points = 20).

Thanks!

[ptc-1368288]

You have two very different questions, like opening two gates. Behind one is the ravine, behind the other is the prairie. What I mean is that this forum is applied rather than a crash course . A project starts by a work sheet on the design board rather than a cartoon work sheet in the blue. Many of your question are already replied in the qs examples.

jmG

[IRstuff]

On 7/12/2009 9:52:17 PM, mj1627 wrote:
>I would like to know if there
>is a method in stats where we
>can compare the two curves
>(data set A & data set B) with
>different number of points?
>
>I would like to know if there
>is a significant difference
>between the data I obtained
>using a model equation (with
>number of points = 150) to the
>experimental data (with number
>of points = 20).
>
>Thanks!

This sounds identical to a question posted on Eng-Tips, except that the number of datapoints is given for the model curve. With that many points, you should be able to do a simple interpolation with the two nearest points to come up with model curve point consistent with your experimental data. This is something that a high school senior should be able to do, and should be a trivial exercise for an engineer.

TTFN,
Eden

[ValeryOchkov]

Why the Math model can not have 20 points. A math model has not points but curves! Or?
Val
http://twt.mpei.ac.ru/ochkov/v_ochkov.htm

[TheodoreM.Bones]

The Student's t-test statistics function may be used without any data manipulation.

[PhilipOakley]

If you have a model, do you have it algebraicallly? this would allow better comparisons because you have (implicitly) better understanding of the error sources (i.e. the model is assumed perfect)

Philip Oakley

[TheodoreM.Bones]

No model, just a collection of numerical data here, one much shorter than the first but centered about a common mean.

[RichardJ]

How can a model equation have a number of points? If it's an equation you can calculate as many points as you want, where you want them.

Richard