Let g ≥ 1 be an integer and let A/Q be an abelian variety that is isogenous over Q to a product of g elliptic curves defined over Q, pairwise non-isogenous over Q and each without complex multiplication. For an integer t and a positive real number x, denote by π A (x, t) the number of primes p ≤ x, of good reduction for A, for which the Frobenius trace a 1,p (A) associated to the reduction of A modulo p equals t. Assuming the Generalized Riemann Hypothesis for Dedekind zeta functions, we prove thatThese bounds largely improve upon recent ones obtained for g = 2 by H. Chen, N. Jones, and V. Serban, and may be viewed as generalizations to arbitrary g of the bounds obtained for g = 1 by M.R. Murty, V.K. Murty, and N. Saradha, combined with a refinement in the power of log x by D. Zywina. Under the assumptions stated above, we also prove the existence of a density one set of primes p satisfying |a 1,p (A)| > p 1 3g+1 −ε for any fixed ε > 0.the Frobenius traces defined by the product of non-isogenous elliptic curves defined over Q and having no complex multiplication, as explained below.Let g ≥ 1 be an integer and let A/Q be an abelian variety that is isogenous over Q to a product of g elliptic curves defined over Q, pairwise non-isogenous over Q and each without complex multiplication.