[Orca-users] Question regarding Statistical Calculations...

Hakan Halisçelik (BS SNYB) Hakan.Haliscelik at disbank.com.tr
Thu Feb 24 22:29:10 PST 2005


i have a problem like this
 
i have a peak at 90% in daily graph but when i look at monthly or yearly graph this 90% looks like 60%
 
so what is the problem, ??
 

	Saygılarımla, 
	Hakan Halisçelik, 
	BS SNYB 
	Tel: 0 212 622 18 00/7042 
	hakan.haliscelik at disbank.com.tr 

 

________________________________

From: orca-users-bounces+hakan.haliscelik=disbank.com.tr at orcaware.com [mailto:orca-users-bounces+hakan.haliscelik=disbank.com.tr at orcaware.com] On Behalf Of David Michaels
Sent: Thursday, February 24, 2005 8:08 PM
To: Sanford, Klaus G.
Cc: 'orca-users at orcaware.com'
Subject: Re: [Orca-users] Question regarding Statistical Calculations...


Sanford, Klaus G. wrote: 

	Dear Sir or Ma'am,

	 

	  What is the statistical definition and value of taking the minimum and maximum values in a data set and taking their average.

	 

	(e.g.  sample size=30 ;  min =8  ; max =79 ; take avg = 8+79 / 2 = 43.5)

	 

	  So 43.5 is the avg of the max and min value but what is the term for this calculation?  Thanks.


I don't think the average of a min and max value has any stastical significance whatsoever.  Perhaps it would be termed the "midpoint" or somesuch.  But I'm not sure what context you're asking this in -- is there something in Orca that relates to your question?  Many orca graphs list a "min, average, max" tuple of data for a particular object, but the 'average' isn't necessarily the average formula you cite above.  The average is calculated over the timescale of the graph in question.  It could be an hour, a day, a week, a month, a quarter, or a year.

For example, look at this graph:

 

As you can see, it's a weekly graph.  Look at the big spike in disk activity on Wednesday, around noon (interesting corresponding spike in network activity - someone was probably streaming some big data file across NFS).  Even though the 'max' for this graph for Disk is 4000, and the min is 0, note that the average is not 2000, but 238.

I'm not positive, but I believe this average is just that -- a statistical average.  The sum of values divided by the quantity of values.  There are plenty of other statistical evaluations you can impose and present in the graph, such as mean, standard deviation, variance, etc.  Maybe put that on the slate of discussion for Orca 3.0. ;)

--Dragon


-------------- next part --------------
An HTML attachment was scrubbed...
URL: </pipermail/orca-users/attachments/20050225/04527b31/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: o_npd-blizzard_gauge_state_c_per_2, __state_D_per_2,	__state_N_per_2, __state_t_per_2, __state_n_per_2,	__state_s_per_2, __state_r_per_2, __state_k_per_2,	__state_m_per_2, __state_d_per_2, __state_i_per_2-weekly.png
Type: image/png
Size: 18181 bytes
Desc: o_npd-blizzard_gauge_state_c_per_2, __state_D_per_2,	__state_N_per_2, __state_t_per_2, __state_n_per_2,	__state_s_per_2, __state_r_per_2, __state_k_per_2,	__state_m_per_2, __state_d_per_2, __state_i_per_2-weekly.png
URL: </pipermail/orca-users/attachments/20050225/04527b31/attachment.png>


More information about the Orca-users mailing list