For every 4×4 real symmetric non-singular matrix A,there exist a positive integer p such that $\exp(pA)-I$ is positive definite.Is it correct?
I know one result which states that if u is eigenvalue of any transformation T and $P(x)\in K[x]$ then p(u) is eigenvalue of p(T).We know that eigenvalue of $exp(pA)-I$ is $\exp(pu)-1$, if u is eigenvalue of A.Here u Can be positive or negative.If it is negative then for positive p $\exp(pu)-1 \lt 0$.Hence $\exp(pA)-I$ is not positive definite.Is my approach correct?