sydneyguy
asked on
vb c++ or c# working with avg of clutter data and trying to find code to find correct avg
Hello thanks for stopping by
looking for some intelligent code that can determine the correct avg of cluster data
have a series of data for sales figures and am looking for alternate ways to come up with the true average of the month long sales for a company,
so have included two images to show what i mean.
one chart shows very high top values well outside the cluster, if i dropped off the top 10% and the bottom 10% to remove the excess top and bottom extremes, this may not get all the high values or as in the 2nd one will not make any diff to the bottom but miss the ones on the top.
am looking at doing some form of center x+ and x- and counting the values amounts till the numbers become greater than a certain increase them drop these ones of.
both + and - to give me a true average of the bulk of the number clutter and as you can see on the pictures the clutter will vary a lot depending on the month and sales figures.
i would still like to drop off the extremes just like test scores. take of the best take off the worst then the avg is some were in side
any ideas would be helpful
regards and thanks
garry
chart1.png
chart2.png
looking for some intelligent code that can determine the correct avg of cluster data
have a series of data for sales figures and am looking for alternate ways to come up with the true average of the month long sales for a company,
so have included two images to show what i mean.
one chart shows very high top values well outside the cluster, if i dropped off the top 10% and the bottom 10% to remove the excess top and bottom extremes, this may not get all the high values or as in the 2nd one will not make any diff to the bottom but miss the ones on the top.
am looking at doing some form of center x+ and x- and counting the values amounts till the numbers become greater than a certain increase them drop these ones of.
both + and - to give me a true average of the bulk of the number clutter and as you can see on the pictures the clutter will vary a lot depending on the month and sales figures.
i would still like to drop off the extremes just like test scores. take of the best take off the worst then the avg is some were in side
any ideas would be helpful
regards and thanks
garry
chart1.png
chart2.png
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
thanks for your help
ASKER
this may vary for every new set of data.
i am trying to put together an aprox value, its can never be exactly correct for every new test data will have different results,
but i am trying to take out the very high and the very low example and take the clutter only into account. this figure could be varied say 80% around the clutter average or even 90% or 60% once i figure out have to identify the clutter then i can take an average of this figure from here i can play with the results
am reading through the data examples that yo have sent me thanks for that
garry