A training-based two-step approach which addresses carrier frequency offset (CFO) and channel frequency response (CFR) estimation in single-input multiple-output (SIMO)-filter bank multicarrier (FBMC) systems based on offset quadrature amplitude modulation (OQAM), is presented. A coarse CFO estimator which does not require the channel knowledge, is investigated in first step. For the combined estimation of fine CFO and CFR in second step, a least squares (LS) estimator based on the conventional interference approximation method (IAM), is proposed. The Cramer-Rao lower bound (CRLB) is derived for the proposed estimator. Simulation results show that the proposed estimator in second step achieves CRLB in the lower signal to noise ratio (SNR) regime. The bit error rate (BER) performance of the proposed approach is also numerically evaluated.