||Asset pricing explaining link between risk and return of an asset remains attractive topic since its inception. Firstly, it has become a crucial tool for corporate managers who wish to estimate cost of capital for an investment and thus determine the demanded risk premium. Secondly, if we assume normative nature of asset pricing model, it could guide an investor to an miss-priced asset and consequently exploit a trading opportunity. We argue that classical covariance-based methods do not adequately capture risk of an asset and thus cannot fully capture the reality of formation of asset returns due to possible asymmetric dependence structures previously documented by empirical literature. It is expected that assets that are highly correlated with representative economic variable in their lower quantiles should require higher risk premium than implied by classical methods, because they do not constitute as good diversification opportunity. Moreover, risk varies with different expectations in various investment horizons. Using quantile cross-spectral analysis proposed by Baruník and Kley (2015), we aim to bridge two literatures: one which utilizes frequency domain analysis for asset pricing, and one which considers asymmetry features of the asset returns. This approach will better mirror behavior of economic agents and their utility function.