completing a bell curve

I am trying to take the area under the curve (area under bell curve) but my data is only 75% complete. Since bell curves are symmetric, is there any way that I can have Matlab complete the curve (based on the data already given to make most of the plot) so that I can take the area under the curve? Thanks!

Answer 1

[Wayne King]

Hi Tara, when you say bell curve, do you mean a Gaussian pdf? If so you can get that information exactly if you know the paramaters (mu and sigma). If you are just talking about a smooth curve, can you use interp1 ? Not sure how your 75% are distributed along the curve.

[Walter Roberson]

If it is like a grading curve, with fixed upper and lower bounds, then gaussian would not be appropriate. A beta distribution might be, though:

http://en.wikipedia.org/wiki/Beta_distribution

Answer 2

[Tara]

Thanks for the answers and suggestions! Basically, I have some data that consistently makes a bell curve (HPLC data). My most recent data collection gave me almost all of the bell curve, except for the last 20% of it. If you look at the wiki beta distribution link, I have all of the teal curve, except from 0.5 down to 1. I want to fill in that data. Do I need to do an extrapolation? Thanks!

[Walter Roberson]

The "teal" curve is the one with the peak before 0.5 ? The one that has only a small tail from 0.5 up to 1? That does not seem to be compatible with your data being "75% complete" ?

[Walter Roberson]

Perhaps you could post a link to a copy of what you _do_ have ?

Answer 3

[UJJWAL]

Hi Tara,

I still have some doubts about your question but I think what you want to ask is that you have an incomplete bell curve from your data collection and you want to complete it. So my suggestion is to extrapolate it since the data collection cannot complete it.....

Hope it Helps. for details reply back

Happy to Help

UJJWAL