RMO Mock Paper 3 · IMOolympiad.com · Original practice

6 questions · 180 minutes · Written proofs

Move from finding an answer to explaining a complete argument. State your assumptions, justify the main step and account for every case.

Original independent practice. Use the suggested time for a full attempt; keep hints and solutions closed. Questions are not official RMO questions. No selection or score prediction is implied.

Question 1

Find every function f:Z→Zf:\mathbb Z\to\mathbb Z satisfying f(x+y)+f(x−y)=2f(x)+6y2f(x+y)+f(x-y)=2f(x)+6y^2 for all integers x,yx,y, with f(0)=0f(0)=0 and f(1)=4f(1)=4.

Hint 1

Use y=1 to obtain a second-difference recurrence.

Hint 2

Subtract the candidate quadratic 3x2+x3x^2+x, including for negative indices.

Worked solution 1

Start with y=1y=1. The given equation becomes f(x+1)−2f(x)+f(x−1)=6.f(x+1)-2f(x)+f(x-1)=6. To find a function with this property, note that (x+1)2−2x2+(x−1)2=2.(x+1)^2-2x^2+(x-1)^2=2. Therefore 3x23x^2 gives the required value 6. Adding a linear expression does not change it, because the corresponding difference of Ax+BAx+B is zero. The values f(0)=0f(0)=0 and f(1)=4f(1)=4 suggest the candidate q(x)=3x2+xq(x)=3x^2+x: it has q(0)=0q(0)=0, q(1)=4q(1)=4, and the required difference.

We must prove that no other function works. Define g(x)=f(x)−q(x)g(x)=f(x)-q(x). Subtracting the difference equations for ff and qq gives g(x+1)=2g(x)−g(x−1),g(0)=g(1)=0.g(x+1)=2g(x)-g(x-1),\qquad g(0)=g(1)=0. It follows that g(2)=2g(1)−g(0)=0g(2)=2g(1)-g(0)=0. Whenever two consecutive values are zero, the next is zero by the same rule. Induction therefore gives g(x)=0g(x)=0 for every nonnegative integer xx.

For negative integers, rearrange the rule as g(x−1)=2g(x)−g(x+1).g(x-1)=2g(x)-g(x+1). With x=0x=0, this gives g(−1)=2g(0)−g(1)=0g(-1)=2g(0)-g(1)=0. With x=−1x=-1, it gives g(−2)=0g(-2)=0, and repeating proves that every negative value is also zero. Hence f(x)=3x2+xf(x)=3x^2+x for all integers xx.

The equation with y=1y=1 was only a necessary condition, so we finish by checking the full equation. For all integers x,yx,y, 3(x+y)2+(x+y)+3(x−y)2+(x−y)=6x2+2x+6y2=2(3x2+x)+6y2.3(x+y)^2+(x+y)+3(x-y)^2+(x-y)=6x^2+2x+6y^2=2(3x^2+x)+6y^2. Thus the candidate satisfies every condition and is the unique answer.

Answer / conclusion: f(x)=3x²+x for every integer x.

Review the idea: Solving functional equations · Second order recurrences

Question 2

Determine all integer triples (x,y,z)(x,y,z) satisfying x2+y2=6z2x^2+y^2=6z^2.

Hint 1

Squares modulo 3 are 0 and 1.

Hint 2

If a nonzero solution exists, divide all three entries by 3 and compare sizes.

Worked solution 2

Modulo 3, the equation gives x2+y2≡0x^2+y^2\equiv0. As each square is 0 or 1, both xx and yy are divisible by 3. Write x=3u,y=3vx=3u,y=3v. Then 3(u2+v2)=2z23(u^2+v^2)=2z^2, so 3∣z3\mid z as well. Put z=3wz=3w; substitution gives u2+v2=6w2u^2+v^2=6w^2, another integer solution.

If a nonzero solution existed, choose one minimising the positive integer ∣x∣+∣y∣+∣z∣|x|+|y|+|z|. The construction produces a nonzero solution with one third that sum, a contradiction. Therefore only (0,0,0)(0,0,0) is possible; it plainly satisfies the equation.

Answer / conclusion: Only (0,0,0).

Review the idea: Remainders · Proof methods

Question 3

In a right triangle ABCABC with ∠A=90∘\angle A=90^\circ, let DD be the foot of the altitude to BCBC. Let E,FE,F be the incentres of triangles ABD,ACDABD,ACD, and let rr be the inradius of ABCABC. Prove that EF=2 rEF=\sqrt2\,r.

Incentres in the two altitude trianglesABCDEFOriginal construction; the proof does not rely on the drawing.

Hint 1

The two smaller triangles are similar to the original triangle.

Hint 2

Use coordinates with D at the origin, BC horizontal and DA vertical.

Worked solution 3

Write a=BCa=BC, b=CAb=CA, c=ABc=AB, and let r1,r2r_1,r_2 be the inradii of triangles ABD,ACDABD,ACD, respectively. Triangle ABDABD and triangle ABCABC are both right triangles and share the acute angle at BB, so they are similar by the angle-angle criterion. Their hypotenuses are AB=cAB=c and BC=aBC=a, so the scale factor is c/ac/a. A similarity scales all lengths, including an inradius, by the same factor. Thus r1=cr/ar_1=cr/a. Likewise, ACDACD and ABCABC share their acute angle at CC and are right triangles, giving r2=br/ar_2=br/a.

Choose perpendicular coordinate axes through DD, with DBDB on the negative horizontal ray, DCDC on the positive horizontal ray, and DADA on the positive vertical ray. The incentre of a right triangle lies inside the triangle and is at distance equal to its inradius from each leg. Therefore E=(−r1,r1)E=(-r_1,r_1) and F=(r2,r2)F=(r_2,r_2).

The horizontal separation of E,FE,F is r1+r2r_1+r_2; their vertical separation is ∣r2−r1∣|r_2-r_1|. Applying Pythagoras to these perpendicular separations gives EF2=(r1+r2)2+(r2−r1)2=2(r12+r22).EF^2=(r_1+r_2)^2+(r_2-r_1)^2=2(r_1^2+r_2^2). Substitute the two inradius formulas and then use b2+c2=a2b^2+c^2=a^2, Pythagoras in ABCABC: EF2=2r2c2+b2a2=2r2.EF^2=2r^2\frac{c^2+b^2}{a^2}=2r^2. Since lengths are positive, taking square roots yields EF=2 rEF=\sqrt2\,r.

Answer / conclusion: EF=√2 r.

Review the idea: Similar triangles · Pythagoras and stewart · Parallel lines and angle bisectors

Question 4

A subset SS of {1,2,…,30}\{1,2,\ldots,30\} contains no two distinct elements whose sum is 31 or 32. Determine the largest possible size of SS, and count all subsets of that largest size.

Hint 1

Order the numbers as 1,30,2,29,3,28,…,15,161,30,2,29,3,28,\ldots,15,16.

Hint 2

The forbidden pairs become consecutive positions in a path of length 30. Count ways to choose 15 nonconsecutive positions.

Worked solution 4

Arrange the numbers in the order 1,30,2,29,3,28,…,15,16.1,30,2,29,3,28,\ldots,15,16. For 1≤k≤151\le k\le15, positions 2k−1,2k2k-1,2k contain k,31−kk,31-k, whose sum is 31. These are all pairs of distinct numbers in the range with sum 31. For 2≤k≤152\le k\le15, positions 2k−2,2k−12k-2,2k-1 contain 32−k,k32-k,k, whose sum is 32. These are all the distinct pairs with sum 32: 1+311+31 uses a number outside the range, and 16+1616+16 does not use distinct elements. Thus two numbers are forbidden together exactly when their positions are consecutive in this list.

Pair positions (1,2),(3,4),…,(29,30)(1,2),(3,4),\ldots,(29,30). At most one position from each pair can be selected, so there can be at most 15 selected numbers. Choosing all odd positions attains 15, since no two are consecutive. Hence the greatest size is 15.

It remains to count all selections of that size. Write the selected positions in increasing order as i1,…,i15i_1,\ldots,i_{15}. They satisfy 1≤i1,ik+1≥ik+2,i15≤30.1\le i_1,\qquad i_{k+1}\ge i_k+2,\qquad i_{15}\le30. Before the kk-th selected position, at least k−1k-1 unselected positions are needed to separate it from the preceding selections. Remove these compulsory gaps by defining jk=ik−(k−1)j_k=i_k-(k-1). Then jk+1−jk=(ik+1−ik)−1≥1,j_{k+1}-j_k=(i_{k+1}-i_k)-1\ge1, and j1≥1j_1\ge1, j15=i15−14≤16j_{15}=i_{15}-14\le16. Thus the jkj_k form a choice of 15 distinct numbers from 1,…,161,\ldots,16.

This correspondence is reversible: from any such increasing choice define ik=jk+k−1i_k=j_k+k-1. Then ik+1−ik≥2i_{k+1}-i_k\ge2, i1≥1i_1\ge1, and i15≤30i_{15}\le30, so it gives an allowed selection. There are (1615)=16\binom{16}{15}=16 choices, one for each number omitted from 1,…,161,\ldots,16. Therefore exactly 16 sets attain the maximum size 15.

Answer / conclusion: Maximum size 15; exactly 16 maximum sets.

Review the idea: Counting with bijections · Combinations and binomial coefficients

Question 5

Real numbers a,b,ca,b,c lie in [−1,1][-1,1] and satisfy a+b+c=0a+b+c=0. Find the largest and smallest possible values of abcabc, including every equality case.

Hint 1

When the product is positive and nonzero, exactly one variable is positive.

Hint 2

Write the other two as −u and −v, and use uv≤(u+v)2/4uv\le(u+v)^2/4.

Worked solution 5

If abc>0abc>0, the zero-sum condition forces two variables to be negative and one positive. After relabelling write (a,b,c)=(−u,−v,u+v)(a,b,c)=(-u,-v,u+v), where u,v>0u,v>0 and u+v≤1u+v\le1. Then abc=uv(u+v)≤(u+v)34≤14.abc=uv(u+v)\le\frac{(u+v)^3}{4}\le\frac14. Equality in the first bound requires u=vu=v, and in the second requires u+v=1u+v=1. Hence the maximum 1/41/4 occurs exactly at permutations of (1,−1/2,−1/2)(1,-1/2,-1/2). A zero or negative product cannot improve this positive maximum. Replacing all three variables by their negatives preserves the domain and sum, and negates the product. Therefore the minimum is −1/4-1/4, attained exactly at permutations of (−1,1/2,1/2)(-1,1/2,1/2).

Answer / conclusion: Maximum 1/4 and minimum −1/4, with stated permutations.

Review the idea: Arithmetic geometric and harmonic means · Inequality rules and signs

Question 6

Among positive integers NN having at least three distinct prime divisors and admitting N=x2+y2N=x^2+y^2 with positive coprime integers x,yx,y, determine the least possible NN.

Hint 1

Show that no prime congruent to 3 modulo 4 can divide N.

Hint 2

The two smallest primes congruent to 1 modulo 4 are 5 and 13. Check 130.

Worked solution 6

Suppose a prime p≡3(mod4)p\equiv3\pmod4 divides x2+y2x^2+y^2. Coprimality implies p∤yp\nmid y, since otherwise p∣xp\mid x too. Thus t≡xy−1(modp)t\equiv xy^{-1}\pmod p satisfies t2≡−1t^2\equiv-1. Raising to (p−1)/2(p-1)/2, an odd integer, gives tp−1≡−1(modp)t^{p-1}\equiv-1\pmod p, contradicting Fermat’s little theorem. Hence every odd prime divisor of NN is 1 modulo 4. Also 4∤N4\nmid N: coprime x,yx,y cannot both be even, and squares modulo 4 are 0 or 1.

The three smallest permitted distinct primes are therefore 2,5,132,5,13. Any admissible NN is at least their product, 130. Finally 130=32+112130=3^2+11^2, with gcd⁡(3,11)=1\gcd(3,11)=1, and its prime divisors are exactly 2,5,132,5,13. Thus 130 is attained and is the minimum.

Answer / conclusion: 130 = 3²+11².

Review the idea: Number theory theorems · Unique prime factorisation

After this paper

Record one idea you missed and one proof or calculation you want to improve. Work through the linked lesson, then try the next paper without hints.

Choose another RMO paper · Find a concept or theorem

Format reference: official RMO programme. Paper content is independently authored practice.