You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Calculating impacts for waterfall graphs
- Thread starter Ross F22
- Start date
D
Deleted member 4993
Guest
As posted the problem does not make sense to me.Hi!
I am trying to quantify the impact of changes in x and y for a waterfall graph I am creating. I.E. prior quarter vs current quarter.
The equation is (x-y)\x
Any help is greatly appreciated!
Please post the EXACT problem - verbatim - as it was given to you.
I don’t have a exact problem as it’s for work and not a class. I’ll try to elaborate:
I want to quantify the drivers of change in a ratio that we look at.
so if it was 75% last month and 70% this month. I want to show how it decreased by 5% (I.e did -2% come from changes to x and -3% from changes to y) by having the impacts of the two variables in the equation below:
75%=(x-y)/x
70%=(x-y)/x
75%=(4,000-1,000)/4,000
70%=(5,000-1,500)/5,000
so how many pp of the 5pp decrease came from changes in x and how many from changes in y?
hope this helps
I want to quantify the drivers of change in a ratio that we look at.
so if it was 75% last month and 70% this month. I want to show how it decreased by 5% (I.e did -2% come from changes to x and -3% from changes to y) by having the impacts of the two variables in the equation below:
75%=(x-y)/x
70%=(x-y)/x
75%=(4,000-1,000)/4,000
70%=(5,000-1,500)/5,000
so how many pp of the 5pp decrease came from changes in x and how many from changes in y?
hope this helps
Last edited:
First, what you probably want to do cannot be done exactly if both variables have changed.
Second, you have made it much harder for yourself by framing the problem in terms of percentage points rather than percentage change. I might have some time today to work on the latter problem; I have no interest in working with percentage points unless a simplification I thought up works (and I'm not sure it does).
Third, it is a mistake to worry about how to present your analysis before you have any idea what the analysis even says. Moreover, it is an even bigger mistake to assume that your audience are all mathematically uninformed. Long ago I did something like what you are talking about for a board that had a whole range of talented people on it: some mathematically sophisticated and some not. When the mathematically competent begin to probe, you better hope that at the end they are satisfied or else your work is wasted.
Second, you have made it much harder for yourself by framing the problem in terms of percentage points rather than percentage change. I might have some time today to work on the latter problem; I have no interest in working with percentage points unless a simplification I thought up works (and I'm not sure it does).
Third, it is a mistake to worry about how to present your analysis before you have any idea what the analysis even says. Moreover, it is an even bigger mistake to assume that your audience are all mathematically uninformed. Long ago I did something like what you are talking about for a board that had a whole range of talented people on it: some mathematically sophisticated and some not. When the mathematically competent begin to probe, you better hope that at the end they are satisfied or else your work is wasted.