Re: fitting and extrapolating

Wouldn't you just fit bins 1-6 to line and use the parameters of the line to estimate how many events are in bin 0?

If you have a TF1 you should be able to use the Integral function to get the number in bin 0.

-Art S.

http://root.cern.ch/root/html402/TF1.html#TF1:Integral

Double_t Integral(Double_t a, Double_t b, const Double_t *params, Double_t epsilon)

*-*-*-*-*-*-*-*-*Return Integral of function between a and b*-*-*-*-*-*-*-*

    based on original CERNLIB routine DGAUSS by Sigfried Kolbig     converted to C++ by Rene Brun

/*

This function computes, to an attempted specified accuracy, the value of the integral

displaymath120

Usage:

In any arithmetic expression, this function has the approximate value of the integral I.

a,b

     End-points of integration interval. Note that B may be less than A. params

     Array of function parameters. If 0, use current parameters. epsilon

     Accuracy parameter (see Accuracy).

A.E. Snyder, Group EC                        \!c*p?/
SLAC Mail Stop #95                          ((.   .))
Box 4349                                        |
Stanford, Ca, USA, 94309                      '\|/`
e-mail:snyder_at_slac.stanford.edu                 o
phone:650-926-2701                              _
http://www.slac.stanford.edu/~snyder          BaBar
FAX:650-926-2657                          Collaboration

On Mon, 11 Jan 2010, asaddiqu_at_ualberta.ca wrote:

> Hi,
>
> I have a histogram with 6 bins{0-6}. Bin zero is unknown. Bin 1-5 correspond
> to measured background. The background falls of according to a straight line
> law.
> Does anybody know how to estimate the number of background events in the bin
> zero by fitting bins 0,1,2,3,4,5,6 to a straight line with:
>
> 1) Intercept floating.
> 2) Slope floating
> 3) No. of events in zeroth bin.
>
>
> Regards, Asif
>
Received on Tue Jan 12 2010 - 02:17:41 CET