For any integer n≥2 and two n×n matrices with real
entries A, B that satisfy the equation
A−1+B−1=(A+B)−1
prove that det(A)=det(B).
Does the same conclusion follow for matrices with complex entries?
(Proposed by Zbigniew Skoczylas, Wrocł aw University of Technology)
Solution (official)
Multiplying the equation by (A+B) we get
I=(A+B)(A+B)−1=(A+B)(A−1+B−1)==AA−1+AB−1+BA−1+BB−1=I+AB−1+BA−1+IAB−1+BA−1+I=0.
Let X=AB−1; then A=XB and BA−1=X−1, so we have
X+X−1+I=0; multiplying by (X−I)X,
0=(X−I)X⋅(X+X−1+I)=(X−I)⋅(X2+X+I)=X3−I.
Hence,
X3=I(detX)3=det(X3)=detI=1detX=1detA=det(XB)=detX⋅detB=detB.
In case of complex matrices the statement is false. Let
ω=21(−1+i3). Obviously
ω∈/R and ω3=1, so
0=1+ω+ω2=1+ω+ωˉ.
Let A=I and let B be a diagonal matrix with all entries along
the diagonal equal to either ω or ωˉ=ω2
such a way that det(B)=1 (if n is not divisible by 3 then
one may set B=ωI). Then A−1=I,
B−1=Bˉ. Obviously I+B+Bˉ=0 and
(A+B)−1=(−Bˉ)−1=−B=I+Bˉ=A−1+B−1.
By the choice of A and B, detA=1=detB.
How the field did
contestants scored
318
average (of 10)
7.43
solved (≥ 80%)
63.8%
near-0 (≤ 10%)
7.9%
discrimination
0.43
Score distribution (field cohort)
Computed on contestants with a meaningful total (field cohort); discrimination is the corrected item–total correlation.