In this paper we consider a trading strategy, which consists in buying or selling a financial instrument when the smoothing, non-causal FIR (Final Impulse Response) filter output attains a local minimum or maximum, respectively. Upon this assumption the goal of this paper is to determine the ``best'' non-causal smoothing FIR filters, which provide maximum value of the return from market. The assumed non-causality is obtained by advancing the output signal to compensate for the delay introduced by the {ITALIC a priori} known filter. The best results were obtained for the impulse response given by the Pascal triangle and the family of symmetric power triangles, both for the case of trading with, and without the transaction fee. It was found that the transaction fee dramatically reduces a possible net return, and therefore should not be omitted in market analyzes.

PACS numbers: 89.65.Gh, 02.30.Mv, 02.60.Ed, 89.90.+n