Percentage Decrease

[xxMsJojoxx]

Is the 'percentage decrease' calculated accurately?

	Funding Errors	Percentage Decrease
June	=10/484= 2.1%
August	=6/362=1.7%	-19%
October	=2/384=1%	-40%

 

[Harry_the_cat]

I'd say the last one should be 41% rounded.

 

[JayJay]

You cannot calculate a meaningful Percentage Decrease because your rounding amounts are comparable to the original numbers, particularly for October where rounding effectively doubles the Funding Error. Using unfounded figures we get October Percentage Decrease (2 ÷ 384) ÷ (6 ÷ 362) = 31%, quite far from 41%.

 

[Dr.Peterson]

Is the 'percentage decrease' calculated accurately?

	Funding Errors	Percentage Decrease
June	=10/484= 2.1%
August	=6/362=1.7%	-19%
October	=2/384=1%	-40%

Using exact results from the fractions, I get this:

	Funding Errors​	Actual​	Percentage Decrease​	Actual​
June​	10/484= 2.1%​	2.06612%​
August​	6/362=1.7%​	1.65746%​	-19%​	-19.78%​
October​	2/384=1%​	0.52083%​	-40%​	-68.58%​

How did you get your percentages?

 

[HallsofIvy]

The "decrease", as a quantity goes from A to B, with A> B, is A- B. The "percentage decrease" is (A- B)/A, written as a percentage.

From June to August the percentage changed from 2.06612 to 1.65746, a decrease of 2.06612- 1.65746= 0.40866. The "percentage decrease" is \(\displaystyle \frac{0.40866}{2.06612}= 0.19779\). That is rounded to 19% (I would have rounded to 20%).

 

[xxMsJojoxx]

Using exact results from the fractions, I get this:

	Funding Errors​	Actual​	Percentage Decrease​	Actual​
June​	10/484= 2.1%​	2.06612%​
August​	6/362=1.7%​	1.65746%​	-19%​	-19.78%​
October​	2/384=1%​	0.52083%​	-40%​	-68.58%​

How did you get your percentages?

Thank you, Dr. Peterson.

I guess I just rounded the October funding errors to the nearest whole percentage, and then calculated the percentage decrease using 1.7% from August to compare with the 1% in October. (I didn't realize the rounding would affect the percentage increase by 28+%.

For August's Percentage decrease:
(0.017/0.021)-1≈ -0.19

For October's Percentage decrease:
(0.01/0.017)-1≈ -0.4

I'm not sure if this is a mathematical question -- but do you think the percentage decrease figures (the accurate figures that you provided) is meaningful, to describe the reduction in funding errors?

 

[Dr.Peterson]

If you'd rounded the October number to 0.5 rather than 1 (an error of 50%!) things would have been closer.

The 70% reduction appears to be significant, based only on what I see. Since I don't even know what a "funding error" is, I can't say anything definitive.