Standard Deviation

Michael C. Lee mclsys@home.com
Fri, 18 Sep 1998 17:02:52 -0700


There is a QTP technique in the June 1991 Cognos Support Technical Newsletter
as
follows:

The following technique calculates the mean and standard deviation of an
employee's
billings in one pass using QTP. It uses the formula:S=sqrt((Q-T*M)/(N-1))
Where:    Q is the sum of the squares of the billings;
                T is the sum of the billings;
                N is the count of the billings;
                M is T / N (e.g. the average of the billings); and
                S is the standard deviation of the billings.

> RUN mean_stdev2
> REQUEST mean_stdev
> ACCESS billings
>     SORT ON employee-no OF billings
> TEMPORARY bill-count INT SIZE 4
>     ITEM bill-count COUNT RESET AT &
>           employee-no OF billings
> TEMPORARY total-employee FLOAT SIZE *
>      ITEM total-employee &
>          SUBTOTAL billing of billings &
>           RESET AT employee-no OF billings
>
> TEMPORARY sq-bill
>      ITEM sq-bill = billing^2
>
> TEMPORARY total-sq-employee FLOAT SIZE 8
>      ITEM total-sq-employee SUBTOTAL sq-bill &
>          RESET AT employee-no OF billings
>
> TEMPORARY mean FLOAT SIZE 8
>      ITEM mean = total-employee / bill-count
>
> TEMPORARY std-dev FLOAT SIZE 8
>     ITEM std-dev = 0 if bill-count = 0 &
>     ELSE ((total-sq-employee - mean * total-employee) / &
>                    (bill-count - 1)) ^0.5
> SUBFILE results2 AT employee-no OF billings KEEP &
>         INCLUDE employee OF billings, std-dev, bill-count
>
> BUILD


I haven't tried this myself but the techniques in the newsletters were pretty
good and were
vetted by Cognos personnel including Mr. Deskin.

Regards,

Michael Lee
MCL Systems Inc.

p.s. I hope you have QTP.



Calvin.Dimsha@gsle.gensig.com wrote:

> We're using Quiz 7.29.C8 on a HP3000.
>
> I need to calculate the standard deviation of 5 arguments and it doesn't
> look like Quiz does this out of the box.  Ideally there would be something
> like this:
>
> DEFINE X = STDEV(0,5,0,266,0) which = 118.4196
>
> Has anyone developed a work-around that accomplishes this and that they
> would be willing to share?
>
> Calvin Dimsha
> GS Laboratory Equipment
> Asheville, NC
> calvin.dimsha@gsle.gensig.com
> = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
> Subscribe: "subscribe powerh-l" in message body to majordomo@lists.swau.edu
> Unsubscribe: "unsubscribe powerh-l" in message to majordomo@lists.swau.edu
> powerh-l@lists.swau.edu is gatewayed one-way to bit.listserv.powerh-l
> This list is closed, thus to post to the list, you must be a subscriber.



= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Subscribe: "subscribe powerh-l" in message body to majordomo@lists.swau.edu
Unsubscribe: "unsubscribe powerh-l" in message to majordomo@lists.swau.edu
powerh-l@lists.swau.edu is gatewayed one-way to bit.listserv.powerh-l
This list is closed, thus to post to the list, you must be a subscriber.