Math Prize 2009 Solutions

Download as pdf or txt
Download as pdf or txt
You are on page 1of 17

2009 Solutions

Problem 1 How many ordered pairs of integers (x, y) are there such that

0 < |xy| < 36?

Answer: 524.
Solution: We will first assume that x and y are positive. Either x < y,
x = y, or x > y.
If x = y, we get the inequality x2 < 36. So x is 1, 2, 3, 4, or 5. We have
5 solutions: (1, 1), (2, 2), (3, 3), (4, 4), and (5, 5).
Next suppose that x < y. Then again we get x2 < xy < 36. So x is 1,
2, 3, 4, or 5. If x = 1, then y can go from 2 through 35: 34 solutions. If
x = 2, then y can go from 3 through 17: 15 solutions. If x = 3, then y can
go from 4 through 11: 8 solutions. If x = 4, then y can go from 5 through
8: 4 solutions. If x = 5, then y can go from 6 through 7: 2 solutions. So the
total number of solutions in this case is 63.
If x > y, then by symmetry there are also 63 solutions.
So adding all three cases gives 5 + 63 + 63 = 131.
The analysis so far was under the assumption that x and y are positive.
By symmetry, there are the same number of solutions for all four choices
of signs (positive-positive, positive-negative, negative-positive, and negative-
negative). So the total number of solutions is 4 · 131 = 524 .

Problem 2 If a, b, c, d, and e are constants such that every x > 0 satisfies

5x4 − 8x3 + 2x2 + 4x + 7 b c d e


4
=a+ + 2
+ 3
+ ,
(x + 2) x + 2 (x + 2) (x + 2) (x + 2)4

then what is the value of a + b + c + d + e?


Answer: 18.

1
AT Math Prize for Girls 2009 Solutions

Solution: If we multiply the equation by (x + 2)4 , we will get polynomials


on both sides. Because the two polynomials are equal for all x > 0 (an infinite
number of values), they must be equal for all x. Dividing by (x + 2)4 , we
see that our original equation must be true for all x 6= −2. Note that if we
plug in x = −1, the right side becomes a + b + c + d + e, which is what we
want. If we plug in x = −1 on the left side, we get

a + b + c + d + e = 5(−1)4 − 8(−1)3 + 2(−1)2 + 4(−1) + 7


=5+8+2−4+7
= 18 .

Problem 3 The Fibonacci numbers are defined recursively by the equation

Fn = Fn−1 + Fn−2

for every integer n ≥ 2, with initial values F0 = 0 and F1 = 1. Let Gn = F3n


be every third Fibonacci number. There are constants a and b such that
every integer n ≥ 2 satisfies

Gn = aGn−1 + bGn−2 .

Compute the ordered pair (a, b).


Answer: (4, 1).
Solution: Besides F0 = 0 and F1 = 1, we know that F2 = 1, F3 = 2,
F4 = 3, F5 = 5, F6 = 8, F7 = 13, F8 = 21, and F9 = 34. Thus we have
G0 = 0, G1 = 2, G2 = 8, and G3 = 34. Plugging n = 2 into the equation
Gn = aGn−1 + bGn−2 , we get 8 = 2a, so a = 4. Plugging in n = 3, we get
34 = 8a + 2b, so b = 1. Hence the ordered pair is (4, 1) .
Alternative Solution: (due to Kouichi Nakagawa) Applying the Fi-

2
AT Math Prize for Girls 2009 Solutions

bonacci recurrence many times, we have


Gn = F3n
= F3n−1 + F3n−2
= 2F3n−2 + F3n−3
= 3F3n−3 + 2F3n−4
= 3F3n−3 + F3n−4 + F3n−5 + F3n−6
= 4F3n−3 + F3n−6
= 4Gn−1 + Gn−2 .

So (a, b) = (4, 1) .

Problem 4 The admission fee for an exhibition is $25 per adult and $12 per
child. Last Tuesday, the exhibition collected $1950 in admission fees from at
least one adult and at least one child. Of all the possible ratios of adults to
children at the exhibition last Tuesday, which one is closest to 1? Express
your answer as a fraction in reduced form.
27
Answer: .
25
Solution: Let a be the number of adults and c be the number of children.
Then we get the condition
25a + 12c = 1950 = 25 · 78.
Rearranging gives
12c = 25(78 − a).
So c must be a multiple of 25.
We will make a chart of the multiples of 25. Note that as c goes up by 25,
a goes down by 12.
a c
66 25
54 50
42 75
30 100
18 125
6 150

54 27
By inspection, the ratio of adults to children closest to 1 is 50
, or .
25

3
AT Math Prize for Girls 2009 Solutions

Problem 5 The figure below shows two parallel lines, ` and m, that are
distance 12 apart:
B
m

`
A

A circle is tangent to line ` at point A. Another circle is tangent to line m at


point B. The two circles are congruent and tangent to each other as shown.
The distance between A and B is 13. What is the radius of each circle?
Express your answer as a fraction in reduced form.
169
Answer: .
48
Solution:
B
m

P Y

`
A X

As in the figure above, draw the line perpendicular to line ` passing


through point B. Let X be the intersection point of this line√ and `. By the
Pythagorean Theorem applied to 4AXB, we have AX = 132 − 122 = 5.
Let P be the center of the circle passing through A and let Q be the center
of the circle passing through B. Let r be the radius of each circle. Draw the
line parallel to ` passing through P . Let Y be the intersection point of this
line and XB.
Due to the right angles, AXY P is a rectangle and P Y Q is a right triangle.
So P Y = 5 and XY = r. Hence QY = 12 − 2r. Also P Q = 2r. By the

4
AT Math Prize for Girls 2009 Solutions

Pythagorean Theorem, we get the equation

52 + (12 − 2r)2 = (2r)2 .

Expanding both sides, we get

25 + 144 − 48r + 4r2 = 4r2 .

The quadratic terms cancel:

25 + 144 − 48r = 0.
169
Solving for r gives r = .
48

Problem 6 Consider a fair coin and a fair 6-sided die. The die begins with
the number 1 face up. A step starts with a toss of the coin: if the coin
comes out heads, we roll the die; otherwise (if the coin comes out tails), we
do nothing else in this step. After 5 such steps, what is the probability that
the number 1 is face up on the die? Express your answer as a fraction in
reduced form.
37
Answer: .
192
Solution: If all the coin tosses come out tails, then the die will remain at 1.
The probability that all 5 tosses will be tails is 215 = 32 1
.
Otherwise, if at least one of the tosses comes out heads, then at the
end the die will be in a random position, because each roll has an equal
probability of showing any particular number. The probability that at least
1 31
one toss is heads is 1 − 32 , or 32 . Given that condition, the probability that
the die will be 1 is 16 . So the probability that at least one toss is heads and
the die ends at 1 is 31
32 6
· 1 = 192
31
.
Hence, adding the two cases together, we get that the total probability is
1 31 37
32
+ 192 , which evaluates to .
192

Problem 7 Compute the value of the expression

20094 − 4 × 20074 + 6 × 20054 − 4 × 20034 + 20014 .

Answer: 384.

5
AT Math Prize for Girls 2009 Solutions

Solution: Let f (x) = (2x + 1)4 . Then our problem is to evaluate


f (1004) − 4f (1003) + 6f (1002) − 4f (1001) + f (1000).
That expression is the fourth finite difference of f evaluated at 1000.
Note that f itself is a quartic polynomial with leading coefficient 16. So
the first difference of f is a cubic with leading coefficient 64. The second
difference is a quadratic with leading coefficient 192. The third difference is
a linear function with leading coefficient 384. The fourth difference is the
constant 384 .
Alternative Solution: Let x = 2005. Then our expression becomes
(x + 4)4 − 4(x + 2)4 + 6x4 − 4(x − 2)4 + (x − 4)4 .
We can simplify by first applying the Binomial Theorem, and then adding
the first and fifth terms together and adding the second and fourth terms
together. We get that the expression equals
2 x4 + 6 · 42 x2 + 44 − 4 · 2 x4 + 6 · 22 x2 + 24 + 6 · x4 .
 

The x4 and x2 terms conveniently cancel out. So we are left with


2 · 44 − 8 · 24 = 512 − 128 = 384 .

Problem 8 Which point on the circle (x − 11)2 + (y − 13)2 = 116 is farthest


from the point (41, 25)? Express your answer as an ordered pair.
Answer: (1, 9).
Solution:
y
A

B
C

6
AT Math Prize for Girls 2009 Solutions

In the figure above, let A be the point (41, 25). Let B = (11, 13) be the
←→
center of the circle. The line AB intersects the circle in two points. Let
C be√ the intersection point that is farther from
√ A. The radius of the circle
is 2 29. The distance between A and B is 6 29.
√ By the triangle inequality, every point on the circle is at most distance
8 29 from A. The point C achieves that distance, so it is the point we are
looking for.
√ √ −−→ −→
Because BC = 2 29 and BA = 6 29, the vector BC is 31 of vector BA.
−→ −−→
Because the vector BA is (30, 12), the vector CB is (10, 4). Hence the coor-
dinates of C are (1, 9) .

Problem 9 The figure below is a 4 × 4 grid of points.

Each pair of horizontally adjacent or vertically adjacent points are distance 1


apart. In the plane of this grid, how many circles of radius 1 pass through
exactly two of these grid points?
Answer: 52.
Solution: For each pair of horizontally adjacent points, there are two circles
that go through them and none of the other grid points. This claim can be
verified by looking at the intersection of the two circles of radius 1 centered
at those two points; the two intersection points are the centers of the two
circles that go through the two points. The number of horizontally adjacent
pairs is 4 · 3 = 12. So there are 24 circles in this case.
The same is true for vertically adjacent points. So we get another 24
circles.
Finally, we can draw a circle at each corner point. That’s 4 more circles.
The total is 24 + 24 + 4 = 52 .

7
AT Math Prize for Girls 2009 Solutions
√ 103 √ 103
Problem 10 When the integer 3+5 − 3−5 is divided by 9,
what is the remainder?
Answer: 1.
Solution: By the Binomial Theorem, we have
√ 103 √ 103 100 √
   

103 102 103 101
5+ 3 = 5 + 103 · 5 3+ 5 ·3+ 5 ·3 3
2 3
√ 103
+ · · · + 103 · 5 · 351 + 3 .

Similarly, we have
√ 103 √ 103 100 √
   

103 102 103 101
5− 3 = 5 − 103 · 5 3+ 5 ·3− 5 ·3 3
2 3
√ 103
+ · · · + 103 · 5 · 351 − 3 .

Adding, we get
√ 103  √ 103
 

103 103 101
5+ 3 + 5− 3 =2·5 +2 5 ·3
2
+ · · · + 2 · 103 · 5 · 351 .

On the right side, every term besides the first two are multiples of 9. So
modulo 9, we have
√ 103  √ 103
 

103 103 101
5+ 3 + 5− 3 ≡2·5 +2 5 · 3 (mod 9).
2

Because 53 = 125 ≡ −1 (mod 9), we have 56 ≡ 1 (mod 9). (We could also
see that by Euler’s totient theorem.) Thus 5103 = 5 · 5102 ≡ 5 (mod 9). Also
103 103 101

2
is divisible by 3, so 2 2 5 · 3 is divisible by 9. Thus our total mod
9 is  
103 103 101
2·5 +2 5 · 3 ≡ 2 · 5 + 0 = 10 ≡ 1 (mod 9).
2
So the remainder is 1 .

Problem 11 An arithmetic sequence consists of 200 numbers that are each


at least 10 and at most 100. The sum of the numbers is 10,000. Let L be

8
AT Math Prize for Girls 2009 Solutions

the least possible value of the 50th term and let G be the greatest possible
value of the 50th term. What is the value of G − L? Express your answer as
a fraction in reduced form.
8080
Answer: .
199
10,000
Solution: The 200 numbers have sum 10,000, so their average is 200
= 50.
Hence we can represent the sequence in the form

50 − 199d, 50 − 197d, . . . , 50 − d, 50 + d, . . . , 50 + 197d, 50 + 199d.

Because all the terms are at least 10 (in particular, the first and last
terms), we have
199 |d| ≤ 50 − 10 = 40.
Solving for d gives
40
|d| ≤
.
199
The 50th term is 50 − 101d. From the bound above, we have
40 4040
101 |d| ≤ 101 · = .
199 199
Hence the 50th term is between 50 − 4040 199
and 50 + 4040
199
40
. By setting d = 199
we get a sequence that achieves the lower bound and meets all the conditions
of the problem; so L = 50 − 4040199
. By setting d = − 19940
we get a sequence
that achieves the upper bound and meets all the conditions of the problem;
8080
so G = 50 + 4040
199
. Hence G − L = .
199
Note: Because each term is at least 10 and their average is 50, each term
is at most 90. So the condition that each term is at most 100 was redundant.

Problem 12 Jenny places 100 pennies on a table, 30 showing heads and 70


showing tails. She chooses 40 of the pennies at random (all different) and
turns them over. That is, if a chosen penny was showing heads, she turns
it to show tails; if a chosen penny was showing tails, she turns it to show
heads. At the end, what is the expected number (average number) of pennies
showing heads?
Answer: 46.

9
AT Math Prize for Girls 2009 Solutions

Solution: Let’s first focus on a penny that started by showing tails. The
40
probability that it turned to heads is 100 , or 25 . Since there are 70 such
pennies, the average number of pennies that switched from tails to heads is
70 · 25 = 28.
Now let’s focus on a penny that started off as heads. The probability
that it stayed heads is 1 − 25 , or 35 . There are 30 such pennies, so the average
number of pennies that started and ended with heads is 30 · 53 = 18.
The average number of pennies that ended with heads is the sum of the
two averages above: 28 + 18 = 46 .

Problem 13 The figure below shows a right triangle 4ABC.


A

B D C

The legs AB and BC each have length 4. An equilateral triangle 4DEF is


inscribed in 4ABC as shown. Point D is the midpoint
√ of BC. What is the
area of 4DEF ? Express your answer in the form m 3 − n, where m and n
are positive integers.

Answer: 22 3 − 36.
Solution: We will solve the problem with complex numbers. Let’s impose
a coordinate system where D is the origin, C is the real number 2, B is the
real number −2, and A is −2 + 4i. Then F is of the form −2 + bi for some
real number b. Because E is a 60◦ clockwise rotation from F , we have

E = F [cos(−60◦ ) + i sin(−60◦ )]
√ !
1 3
= (−2 + bi) − i
2 2


 
3 1
= −1 + b + b + 3 i.
2 2

10
AT Math Prize for Girls 2009 Solutions

←→
Because E lies on the line AC, its real and imaginary parts must add up
to 2. So we get √
3 1 √
−1 + b + b + 3 = 2.
2 2
Solving for b, we get

6−2 3  √  √  √
b= √ = 3− 3 3 − 1 = 4 3 − 6.
3+1
The side length squared of the equilateral triangle is thus
 √ 2 √
2 2 2
|F | = 2 + b = 4 + 4 3 − 6 = 88 − 48 3.

Hence the area of the equilateral triangle is


√ √
88 − 48 3 3  √ √ √
= 22 − 12 3 3 = 22 3 − 36 .
4

Problem 14 The three roots of the cubic 30x3 − 50x2 + 22x − 1 are distinct
real numbers between 0 and 1. For every nonnegative integer n, let sn be the
sum of the nth powers of these three roots. What is the value of the infinite
series
s0 + s1 + s2 + s3 + . . . ?

Answer: 12.
Solution: Let p, q, and r be the roots. Then sn = pn + q n + rn . When we
add over all n, we need to find the sum of pn (and q n and rn ). By the sum
1
of a geometric series, that is 1−p . So we need to evaluate

1 1 1
+ + .
1−p 1−q 1−r

The numbers 1 − p, 1 − q, and 1 − r are the roots of the cubic 30(1 − x)3 −
50(1 − x)2 +22(1−x)−1. The constant coefficient of that cubic is 30−50+22−
1 = 1. The linear coefficient of that cubic is 30(−3)+50(2)−22 = −12. Thus
1
, 1 , and 1−r
1−p 1−q
1
are the roots of a cubic in the reversed form 1x3 −12x2 +· · · .
So the sum of 1−p1 1
, 1−q 1
, and 1−r is − −12
1
= 12 .

11
AT Math Prize for Girls 2009 Solutions
q
4
Problem 15 Let x = 3
. There is a unique value of y such that 0 < y < x
25
and xx = y y . What is the value of y? Express your answer in the form c ab ,
p
where a and b are relatively prime positive integers and c is a prime number.
r
32
Answer: 3 .
3125
Solution: We can rewrite x as
r  1/3  2/3
3 4 4 2
x= = = .
25 25 5

The equation xx = y y is equivalent to x log x = y log y. We can rewrite x log x


as
 2/3
2 2 2
x log x = · log
5 3 5
 2/3
2 2 5 2
= · · log
5 5 3 5
 5/3
2 5 2
= · log
5 3 5
 5/3  5/3
2 2
= log .
5 5
5/3
By inspecting the right side, we see that y = 25 satisfies the equation
x log x = y log y. Furthermore 0 < y < x. So we have found the unique value
we were looking for. To express it in the form we were asked for, we have
 5/3  5 1/3  1/3 r
2 2 32 3 32
y= = 5
= = .
5 5 3125 3125

Problem 16 Let x be a real number such that the five numbers cos(2πx),
cos(4πx), cos(8πx), cos(16πx), and cos(32πx) are all nonpositive. What is
the smallest possible positive value of x? Express your answer as a fraction
in reduced form.
21
Answer: .
64

12
AT Math Prize for Girls 2009 Solutions

Solution: If 0 < x < 41 , then 0 < 2πx < π2 , which implies cos(2πx) > 0,
which is false. So x ≥ 41 .
If 14 ≤ x < 16 5
, then 2π ≤ 8πx < 2π + π2 , which implies cos(8πx) > 0,
5
which is false. So x ≥ 16 .
If 16 ≤ x < 64 , then 10π ≤ 32πx < 10π + π2 , which implies cos(32πx) > 0,
5 21

which is false. So x ≥ 21 64
.
If x = 64 , then 2πx = 21
21
32
π, 4πx = 21
16
π, 8πx = 2π + 85 π, 16πx = 4π + 54 π,
and 32πx = 10π + 21 π. All have nonpositive cosines. So the smallest possible
21
value of x is .
64
Alternative Solution: Consider the binary (base 2) representation of
the positive number x. Look at the bits (digits) after the radix point (the dot
or the “decimal” point). Let’s number them Bits 1, 2, 3, and so on. Then
cos(2πx) is nonpositive if and only if Bits 1 and 2 are different. (That’s
because cosine is nonpositive in the 2nd and 3rd quadrants.) Similarly, the
five given cosines are nonpositive if and only if Bits 1, 2, 3, 4, 5, and 6
alternate in value. So, to make x as small as possible, we should choose
21
1
x = (.010101)2 . As a fraction, that number is 41 + 16 1
+ 64 , or .
64

Problem 17 Let a, b, c, x, y, and z be real numbers that satisfy the three


equations

13x + by + cz = 0
ax + 23y + cz = 0
ax + by + 42z = 0.

Suppose that a 6= 13 and x 6= 0. What is the value of


13 23 42
+ + ?
a − 13 b − 23 c − 42

Answer: −2.
Solution: We can rewrite the first equation as

(a − 13)x = ax + by + cz.

13
AT Math Prize for Girls 2009 Solutions

Because a 6= 13 and x 6= 0, both sides of the equation are nonzero. Similarly


we can rewrite the second and third equations as
(b − 23)y = ax + by + cz
(c − 42)z = ax + by + cz.
Again, both sides of both equations are nonzero. Solving for x, y, and z, we
get
ax + by + cz
x=
a − 13
ax + by + cz
y=
b − 23
ax + by + cz
z= .
c − 42
Plugging those values back into the expression ax + by + cz, we have
ax + by + cz ax + by + cz ax + by + cz
ax + by + cz = a · +b· +c·
a − 13  b − 23  c − 42
a b c
= (ax + by + cz) + + .
a − 13 b − 23 c − 42
Cancelling the nonzero ax + by + cz on both sides gives
a b c
+ + = 1.
a − 13 b − 23 c − 42
Hence we have
13 23 42  a   b   c 
+ + = −1 + −1 + −1
a − 13 b − 23 c − 42 a − 13 b − 23 c − 42
a b c
= + + −3
a − 13 b − 23 c − 42
=1−3
= −2 .

Problem 18 The value of 21! is 51,090,942,171,abc,440,000, where a, b, and


c are digits. What is the value of 100a + 10b + c?
Answer: 709.

14
AT Math Prize for Girls 2009 Solutions

Solution: The number 21! is divisible by the primes 7, 11, and 13, so it is
divisible by their product 1001. The divisibility test for 1001 is similar to that
for 11, except that we group digits into blocks of 3. (That works because 1001
is 1 more than 1000.) So we know that 000 − 440 + abc − 171 + 942 − 090 + 51
is divisible by 1001. In other words, abc + 292 is divisible by 1001. The only
way that can be is if abc is 1001 − 292 = 709. So a is 7, b is 0, and c is 9.
The answer is 709 .

Problem 19 Let S be a set of 100 points in the plane. The distance between
every pair of points in S is different, with the largest distance being 30. Let
A be one of the points in S, let B be the point in S farthest from A, and let
C be the point in S farthest from B. Let d be the distance between B and C
rounded to the nearest integer. What is the smallest possible value of d?
Answer: 17.
Solution:
Q

P B
A

Let r = BC. Consider the circle above on the right with center B and
radius r. Because C is the point in S farthest from B, every point in S is in
or on that circle.
Let P be the point on that circle closest to point A. Every point in S
is at most distance AB from A and hence (by the triangle inequality) is at
most distance P A + AB from P . But P A + AB = P B = r. So every point
in S is in or on the circle above on the left with center P and radius r.
Consider the lens above that is the intersection of the two disks of radius r
centered at B and P . By our previous work, every point in S is in that
lens. The two points in the lens that are farthest apart are the points Q

15
AT Math Prize for Girls 2009 Solutions

and R. Because √ P BQ and P BR are equilateral √ triangles, we can calculate


that QR =√r 3. So we get the inequality r 3 ≥ 30. Solving for r, we
find r ≥ 10 3 ≥ 17. So d, which is r rounded to the nearest integer, is at
least 17.
To finish, we need a configuration in which d = 17. Let’s temporarily
ignore the condition that all the distances be different. Look at the following
configuration.

√ √
10 3 15 10 3

A
√ √ B
C 5 3 5 3

√ 15 √
10 3 10 3

In this configuration, the points A and C coincide. The set S will consist
of the five points A, B, C, D, and E, plus 95 points near the center. The
largest distance is DE = 30, as desired. The point B is a point in S farthest

from A, and C is a point in S farthest from B. The distance BC is 10 3.
The only thing wrong with that configuration is that some of the distances
are equal. But we can perturb the points A, B, and C a little away from the
rhombus
√ so that all the distances are different and so that BC is less than
10 3 + 0.1. In that case, BC rounded to the nearest integer will still be 17.
Putting both parts together, we see that the smallest possible value of d
is 17 .

Problem 20 Let y0 be chosen randomly from {0, 50}, let y1 be chosen ran-
domly from {40, 60, 80}, let y2 be chosen randomly from {10, 40, 70, 80}, and
let y3 be chosen randomly from {10, 30, 40, 70, 90}. (In each choice, the pos-
sible outcomes are equally likely to occur.) Let P be the unique polynomial

16
AT Math Prize for Girls 2009 Solutions

of degree less than or equal to 3 such that P (0) = y0 , P (1) = y1 , P (2) = y2 ,


and P (3) = y3 . What is the expected value of P (4)?
Answer: 107.
Solution: Let Q be the expected value of the polynomial P . Then Q is also
a polynomial of degree less than or equal to 3. Writing E for expected value,
we have
0 + 50 50
Q(0) = E(P (0)) = E(y0 ) = = = 25.
2 2
Similarly, we have
40 + 60 + 80 180
Q(1) = E(P (1)) = E(y1 ) = = = 60
3 3
10 + 40 + 70 + 80 200
Q(2) = E(P (2)) = E(y2 ) = = = 50
4 4
10 + 30 + 40 + 70 + 90 240
Q(3) = E(P (3)) = E(y3 ) = = = 48.
5 5
We are trying to find Q(4). Because Q has degree less than or equal to 3,
its fourth finite difference is zero. In particular, we have

Q(4) − 4Q(3) + 6Q(2) − 4Q(1) + Q(0) = 0.

Solving for Q(4), we get

Q(4) = 4Q(3) − 6Q(2) + 4Q(1) − Q(0)


= 4 · 48 − 6 · 50 + 4 · 60 − 25
= 192 − 300 + 240 − 25
= 432 − 325
= 107 .

17

You might also like