Hi,
Let's say I have 4 values A, B, C and D. Let's say they each contribute 30%, 30%, 20% and 20% towards some final value X.
So my weighted average formula is:
X = [0.3(A) + 0.3(B) + 0.2(C) + 0.2(D)] / [0.3+0.3+0.2+0.2]
But what if the situation is such that values C and D contribute negatively to the over all value X?
As a sports analogy using a hypothetical situation, let's say I am calculating a Football Quarterback's rating using my own method as follows:
A = Yards per Attempt (I want this to be high, so this is a positive value) - 30% weight
B = Number of Touchdowns (I want this to be high, so this is a positive value) - 30% weight
C = Number of Incomplete passes (I want this number to be low/miniumum and it contributes negatively to the overall rating) - 20% weight
D = Number of Interceptions (Needs to be minimized and contributes negatively to the overall rating) - 20% weight
So for my overall weighted average QB rating can I use this formula?
X = [0.3(A) + 0.3(B) - 0.2(C) - 0.2(D)] / [0.3+0.3+0.2+0.2]
Also does it matter if the values of A, B, C and D have greatly different values? Take some hypothetical numbers in my example:
A = Yards per Attempt = 7
B = Number of TDs = 3
C = Number of incompletions = 15
D = Number of Interceptions = 1
As you can see the numbers vary by quite a bit. With my formula X = [0.3(A) + 0.3(B) - 0.2(C) - 0.2(D)] am I unintentionally giving extra weight to certain values just because they numbers themselves are much larger than the others?
What would be the best way to go about solving such a problem where some values contribute and other take away from the final value, each number having its own contribution in the whole scheme of things?
Thanks,
Maja
Let's say I have 4 values A, B, C and D. Let's say they each contribute 30%, 30%, 20% and 20% towards some final value X.
So my weighted average formula is:
X = [0.3(A) + 0.3(B) + 0.2(C) + 0.2(D)] / [0.3+0.3+0.2+0.2]
But what if the situation is such that values C and D contribute negatively to the over all value X?
As a sports analogy using a hypothetical situation, let's say I am calculating a Football Quarterback's rating using my own method as follows:
A = Yards per Attempt (I want this to be high, so this is a positive value) - 30% weight
B = Number of Touchdowns (I want this to be high, so this is a positive value) - 30% weight
C = Number of Incomplete passes (I want this number to be low/miniumum and it contributes negatively to the overall rating) - 20% weight
D = Number of Interceptions (Needs to be minimized and contributes negatively to the overall rating) - 20% weight
So for my overall weighted average QB rating can I use this formula?
X = [0.3(A) + 0.3(B) - 0.2(C) - 0.2(D)] / [0.3+0.3+0.2+0.2]
Also does it matter if the values of A, B, C and D have greatly different values? Take some hypothetical numbers in my example:
A = Yards per Attempt = 7
B = Number of TDs = 3
C = Number of incompletions = 15
D = Number of Interceptions = 1
As you can see the numbers vary by quite a bit. With my formula X = [0.3(A) + 0.3(B) - 0.2(C) - 0.2(D)] am I unintentionally giving extra weight to certain values just because they numbers themselves are much larger than the others?
What would be the best way to go about solving such a problem where some values contribute and other take away from the final value, each number having its own contribution in the whole scheme of things?
Thanks,
Maja
