[yharmon6614]

I'm sure this will be terribly basic to some of you, but my maths is not great, so its stumped me.

I have two numbers, for example 45 and 75. I need to know what the caculation is to find out the % difference (margin) between two given numbers. (it must end up as a positive %)

Ultimately I need to write it as an excel formula - so referencing the two cells that contain the numbers.

Can anyone help?

[furillo]

Chris, the problem is that your question is slightly ill posed. What you are looking for I suppose is the percent change of the two numbers? For the number you have given it would be

100*(75-45)/45 = 66.66666... %

In other words 75 is 66.6666...% bigger than 45. You need to compute the difference between the numbers and then divide by one number( the "original value") then multiply by 100 to get a percentage.

[yharmon6614]

No - I need to know the percentage margin between the two numbers. I don't know how better to explain it to be honest I don't need to know how much bigger one number is that the other, its the difference (margin) expressed as a percentage that I am interested in.

If its any help - using another spreadsheet I have here, I get a percentage margin of 40%, when I use those two numbers (45 and 75). Trouble is that spreadsheet is written in such a way that I cannot just lift out the formula to use in the other document I am creating.

[furillo]

Then just do
100*(75-45)/75= 40.0 %

No matter what you will be just finding the difference between the two numbers and dividing by one of the numbers.

[addyta.org]

No - I need to know the percentage margin between the two numbers. I don't know how better to explain it to be honest I don't need to know how much bigger one number is that the other, its the difference (margin) expressed as a percentage that I am interested in.

If its any help - using another spreadsheet I have here, I get a percentage margin of 40%, when I use those two numbers (45 and 75). Trouble is that spreadsheet is written in such a way that I cannot just lift out the formula to use in the other document I am creating.

It won't work without a "fudge factor", the previous poster was correct, as you can't get those values to work without one.
Can you give another example, prefereably a couple, do check the fudging is a fixed or variable factor?
Alternatively - and preferably - can you check with the person who initially prepared the spreadsheet?

To be consistent, though, you may have to accept that there may be a negative value if things went down -you may be able to set it to give an absolute (?) number that won't have a negative value, regardless.

[yharmon6614]

Then just do
100*(75-45)/75= 40.0 %

No matter what you will be just finding the difference between the two numbers and dividing by one of the numbers.

That works great, thank you. I knew it would be simple - I just could not see the answer....

[addyta.org]

That works great, thank you. I knew it would be simple - I just could not see the answer....

Huh? That's like saying you need a formula where 1+1=3 and solving it by writing it as 1+1=3 - or am I missing something?

What you need, IMO, is something like (100*(75-45)/45)k=40%, in this case k=0.6 (I think). However it may not be the same in other cases, it may be progressive, proportional, exponentially increasing etc - hence the request for other samples.

[edit] I mean (100(b-a)/a)k=x%, where a is initial value, b is new value, k is the modifier (term?) and x is the resultant.

[Lapsiks]

You're able to copy and paste the formula you know, just copy the cell and paste as special and then select formula. Also works if you don't want to paste the formula but the results.

[furillo]

Huh? That's like saying you need a formula where 1+1=3 and solving it by writing it as 1+1=3 - or am I missing something?

What you need, IMO, is something like (100*(75-45)/45)k=40%, in this case k=0.6 (I think). However it may not be the same in other cases, it may be progressive, proportional, exponentially increasing etc - hence the request for other samples.

[edit] I mean (100(b-a)/a)k=x%, where a is initial value, b is new value, k is the modifier (term?) and x is the resultant.

I don't think that is what he is looking for but without the context its impossible to know. When you are looking at percent difference you need to define a baseline to which you are referencing the change. Without knowing the problem, it could be either the 45 or the 75 but for whatever reason its the 75. Generally these problems are associated with a time event, i.e at time A the value is 75 and at time B the value is 45 therefore the percent difference when moving from time A to time B is 40 %.

In another context, the 75 could be the exact value while the 45 is some approximation. In such caluclations the exact value is what you divide by to find the percentage error (assuming you take the absolute value on the top).

[addyta.org]

Nope, I don't think so - in the actual example given the arithmetic answer is 66.6666% but the percentage given as the answer is 40%, ergo something has been missed out that modifies the correct answer to the one that is given - hence the additional term 'k'. It could be a more complex modifier but without more info', as you said, it isn't possible to tell.

This is where Neeyik would come in with "There's a plane on a conveyor belt ..."

[furillo]

Nope, I don't think so - in the actual example given the arithmetic answer is 66.6666% but the percentage given as the answer is 40%, ergo something has been missed out that modifies the correct answer to the one that is given - hence the additional term 'k'. It could be a more complex modifier but without more info', as you said, it isn't possible to tell.

This is where Neeyik would come in with "There's a plane on a conveyor belt ..."

Gordo, read my second post. I show exactly where the 40% comes from. I don't mean to sound like a dick but I am University math lecturer (although this problem is more high school related). The problem isn't really a math problem at all but more of the wording of the question.

Anyways, as you say its a ridiculous thing to discuss and the convery belt thread has already crossed my mind....

[maliboia]

You're missing the point about margins Gordo.

In Chris' example, we're calculating the margin of 75 on 45, not the percentage increase of 75 on 45.

75 is the unit received, 45 is the unit given. Therefore a margin of 30 in 75 is made, or 40%.

My guess is this is something to do with the margins made on Chris' contractors between the amount he receives from the company that employs them and the amount that they then pay to the contractor.

[addyta.org]

You're missing the point about margins Gordo.

In Chris' example, we're calculating the margin of 75 on 45, not the percentage increase of 75 on 45.

75 is the unit received, 45 is the unit given. Therefore a margin of 30 in 75 is made, or 40%.

My guess is this is something to do with the margins made on Chris' contractors between the amount he receives from the company that employs them and the amount that they then pay to the contractor.

To the two of you above - now I get it, I think, so for 100 received, 60 may be given with a margin of 40? Nice and simple when you explain it like that, thanks.
It may still be on a sliding scale so it may be an idea for Chris to check some other examples - unless he already know the percentage or margin is constant.

Runs away to bed - after midnight here with a 7am start to work [yawn]

[yharmon6614]

Spacely is right - and no, its actually nothing to do with my work - its something personal I am trying to work out.

[furillo]

Yes, Gurner explains it nicely and puts it into a business example that makes perfect sense in the context.