Solving Hard Matrix Prob: A+kB Invertible w/ Integer Entries

  • Thread starter Thread starter barbiemathgurl
  • Start date Start date
  • Tags Tags
    Hard Matrix
barbiemathgurl
Messages
12
Reaction score
0
i just can't figure this out.

given a n x n matrix (with n>1) "A" such that all entries are integers and A is invertible such that A^{-1} also has integer entries. Let B be another matrix with integer coefficients so that:
A+B, A+2B, A+3B, ... A+(n^2)B
Are all invertible with integer entries.

Show that,
A+kB
Is also invertible with integer enties for any integer k.

who the heck do you solve this?
 
Mathematics news on Phys.org
What can you say about the determinants of those matrices? Once you get this, use the fact that det(A+kB), with A,B known and k a variable, is a polynomial in k.
 
barbiemathgurl said:
i just can't figure this out.

given a n x n matrix (with n>1) "A" such that all entries are integers and A is invertible such that A^{-1} also has integer entries. Let B be another matrix with integer coefficients so that:
A+B, A+2B, A+3B, ... A+(n^2)B
Are all invertible with integer entries.

Show that,
A+kB
Is also invertible with integer enties for any integer k.

who the heck do you solve this?

Let M be an arbitrary square invertible matrix whose inverse and itself has integer entires. Then 1=\det (MM^{-1}) = \det(M)\det (M^{-1}) shows that \det (M) = \pm 1 because the determinant of this matrix must be an integer. Now define the function f(x) = \det (A+xB). This is a polyomial of at most n degree. Notice that f(0),f(1),f(2),...,f(n^2) are either 1 \mbox{ or }-1. By the strong form of the Pigeonhole Principle at least n+1 of them are either 1 or -1. Without lose of generality say its 1. That means f(x) must in fact be a constant polynomial because a polynomial of at most n degree cannot produce the same values for n+1 different values. So f(x)=1. That means f(k)=1 for no matter what k. So \det (A+kB)=1. Since the determinant is 1, it must mean the matrix is irreducible with integer coefficients (by the adjoint matrix formula).
 
Kummer, we try not to give complete solutions here, just hints. And incidentally, a slightly easier way to get the last step is to note that f(k)^2 is a polynomial of degreen n^2 which is 1 at n^2+1 points, so must be 1 identically, and so f(k)=+-1. Also it remains to show that all integer matrices with determinant +-1 are invertible with integer entries.
 
StatusX said:
Kummer, we try not to give complete solutions here, just hints.
Okay.

Also it remains to show that all integer matrices with determinant +-1 are invertible with integer entries.
Last line in my first post in paranthesis. I was being sloppy on that last line because I assumed that result was trivial. I should have been more explicit.
 
Insights auto threads is broken atm, so I'm manually creating these for new Insight articles. In Dirac’s Principles of Quantum Mechanics published in 1930 he introduced a “convenient notation” he referred to as a “delta function” which he treated as a continuum analog to the discrete Kronecker delta. The Kronecker delta is simply the indexed components of the identity operator in matrix algebra Source: https://www.physicsforums.com/insights/what-exactly-is-diracs-delta-function/ by...
Fermat's Last Theorem has long been one of the most famous mathematical problems, and is now one of the most famous theorems. It simply states that the equation $$ a^n+b^n=c^n $$ has no solutions with positive integers if ##n>2.## It was named after Pierre de Fermat (1607-1665). The problem itself stems from the book Arithmetica by Diophantus of Alexandria. It gained popularity because Fermat noted in his copy "Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos, et...
I'm interested to know whether the equation $$1 = 2 - \frac{1}{2 - \frac{1}{2 - \cdots}}$$ is true or not. It can be shown easily that if the continued fraction converges, it cannot converge to anything else than 1. It seems that if the continued fraction converges, the convergence is very slow. The apparent slowness of the convergence makes it difficult to estimate the presence of true convergence numerically. At the moment I don't know whether this converges or not.
Back
Top