Hi Karen -
I would suggest that you discuss this issue with your client and
find out how they would like to weight the various components. Quite often
there are factors that need to be accounted for other than the most obvious.
For example a given salesperson might be in charge of getting rid of
overstock any way he/she can, including selling at a loss. You can develop
a linear model using those weights in a non-parametric plus or minus fashion.
Another approach would be to use a multiple regression technique
where you derive the weights based on past performance against a measurable
criterion such as profit per transaction or profit during a given time
frame. This would also result in a linear model.
Research has found that people are very good at deciding which
criteria to use in making judgements but lousy at deciding how to weight
those criteria and lousy at combining them to come up with an overall
judgement of how to rank a given event.
Bernie
---------------------------------------
At 09:36 AM 11/7/2002 -0500, you wrote:
>
>I guess I was sleeping when they taught this... I want to
>know if there is an established business statistics theory
>for what I want to do. A client has a simple salesperson's
>profitability report, and he wants it sorted from highest
>profit margin to lowest. We have an RBase report to do this,
>and the report looks like this:
>
> #Sales Sales$ Avg$ Profit%
> Salesman1 1 150 150 30%
> Salesman2 50 10000 200 25
> Salesman3 40 15000 375 20
>
>Obviously sorting by Profit% is misleading. Salesman1 isn't
>exactly their 'best' salesman! Salesman2 had the highest
>number of sales and the highest profit% (after the loafer #1).
>But Salesman3 had highest total sales$ and average$!
>
>Is there a way to WEIGHT the profit%? And what's the best
>indicator of the 'best' salesman? Is it the Total Sales$,
>the Average$, the number of items sold? I guess if there's
>a formula to weight the profit% then I could ask my client
>what other parameter is the most important to them.
>
>
>
>Karen
>
>
>================================================
>TO SEE MESSAGE POSTING GUIDELINES:
>Send a plain text email to [EMAIL PROTECTED]
>In the message body, put just two words: INTRO rbase-l
>================================================
>TO UNSUBSCRIBE: send a plain text email to [EMAIL PROTECTED]
>In the message body, put just two words: UNSUBSCRIBE rbase-l
>================================================
>TO SEARCH ARCHIVES:
>http://www.mail-archive.com/rbase-l%40sonetmail.com/
>
================================================
TO SEE MESSAGE POSTING GUIDELINES:
Send a plain text email to [EMAIL PROTECTED]
In the message body, put just two words: INTRO rbase-l
================================================
TO UNSUBSCRIBE: send a plain text email to [EMAIL PROTECTED]
In the message body, put just two words: UNSUBSCRIBE rbase-l
================================================
TO SEARCH ARCHIVES:
http://www.mail-archive.com/rbase-l%40sonetmail.com/
