2012 AMC 10A Problem 13
An iterative average of the numbers , , , , and is computed the following way. Arrange the five numbers in some order. Find the mean of the first two numbers, then find the mean of that with the third number, then the mean of that with the fourth number, and finally the mean of that with the fifth number. What is the difference between the largest and smallest possible values that can be obtained using this procedure?
Show solution without answering
Key insight
Each averaging step halves the weight of the earlier numbers, so the last number counts most. Put 5 last for the largest result and 1 last for the smallest.
Solution
First see how much each position "counts", then choose the best and worst orders.
1. Try the order to see the pattern:
- mean of and : ,
- mean of and : ,
- mean of and : ,
- mean of and : .
2. Now the general pattern. Each time we take a mean, the new number gets half the weight, and everything before it is cut in half. Following the five positions through:
- after the last step, the fifth number counts for ,
- the fourth for ,
- the third for ,
- the first and second for each.
These add to , so the result is a weighted average: later numbers count more.
3. Largest result. Put the biggest numbers where they count most, so the order is . As in step 1:
4. Smallest result. Put the smallest numbers where they count most, so the order is :
5. Difference: .
The answer is .
Why this works
Repeated averaging always gives each position a fixed share of the result: counting back from the end. Once you know the shares, the largest result comes from giving the biggest share to the biggest number, and the smallest result from giving it to the smallest number.
Concise solution
A shorter write-up, the way contest solutions are usually written. It may take a different route from the solution above.
Suppose the numbers are entered in the order . Track the coefficient of each one:
Each averaging step halves every earlier weight, so later entries count more. The result is a weighted average with weights .
To maximize, give the biggest weights to the biggest numbers: order gives
To minimize, reverse: order gives
Difference: .
The answer is .
Review in the AoPS books
- Prealgebra, Chapter 13: Data and Statistics
- Introduction to Algebra, Chapter 11: Special Factorizations (§11.2, §11.3, §11.5)
- Introduction to Algebra, Chapter 22: Special Manipulations
Similar problems
Problem © Mathematical Association of America (MAA), American Mathematics Competitions. Reproduced for non-commercial educational use. Solution and commentary are original to this site.