Standard Deviation Curve

[rai_lanc]

To all who can help,

I work for a company who deals with customer dwell time between areas
of a shopping centre.
They are tracking people between 2 points to see how long they are
spending in a particular area.

I can run reports myself, but using a Java Based program that
interrogates a database of customers, and where they have been.

Anyway, the information I get is this :

Sample Count : 727
Mean Dwell Time in minutes : 9.6818
Standard Deviation : 22.9481

I also can see a standard deviation curve on the screen, but I can't
export this to Excel...

My problem is, that I already KNOW the Standard Deviation and the Mean
Dwell Time, but I want to re-produce these curves in Excel.

I can't see a way of plotting a curve when I already know the Standard
Deviation.
I'm new to this, and I'm probably missing something somewhere - I just
don't know what!!!

Any help would be greatly appreciated,

Many thanks,

Rachael

 

[Guest]

You may need to provide more information on what you want in your curve. I
just did Google search for Dwell time curve and saw a lot of options.

 

[James Silverton]

(e-mail address removed) wrote on Thu, 13 Sep 2007
02:26:51 -0700:

rl> I work for a company who deals with customer dwell time
rl> between areas of a shopping centre.
rl> They are tracking people between 2 points to see how long
rl> they are spending in a particular area.

rl> I can run reports myself, but using a Java Based program
rl> that interrogates a database of customers, and where they
rl> have been.

rl> Anyway, the information I get is this :

rl> Sample Count : 727
rl> Mean Dwell Time in minutes : 9.6818
rl> Standard Deviation : 22.9481

rl> I also can see a standard deviation curve on the screen,
rl> but I can't export this to Excel...

rl> My problem is, that I already KNOW the Standard Deviation
rl> and the Mean Dwell Time, but I want to re-produce these
rl> curves in Excel.

If you know a mean and standard deviation, any of the several
methods for producing a normal error curve might do what you
want.....even NORMSDIST tho' it's not my favorite method.
I personally prefer

SQRT(LN(RAND())*(-2))*COS(RAND()*2*PI()) *sigma + mean


James Silverton
Potomac, Maryland

E-mail, with obvious alterations:
not.jim.silverton.at.verizon.not