We begin this thesis with developing a theoretical framework and propose a relevant empirical analysis of the soybean complex prices cointegration relationship in a high-frequency setting. We allow for heterogeneous expectations among traders on the multi-asset price dynamics and characterize the resulting market behavior. We demonstrate that the asset prices autoregressive matrix rank and the speed of reversion towards the long-term equilibrium are related to the market liquidity, unlike the cointegrating vector. Our empirical application to the soybean complex, where we control for volatility, supports our theoretical results when the price idleness of the different assets is properly accounted for. Next we switch to price discovery analysis to investigate the lead-lag relationship between equity and CDS markets within a corporate finance framework. Based on investment grade and high yield firms, we establish a panel framework associated to nine corporate financial characteristics factors related to price volatility, default risk, and company capital structures. Contributing to the ongoing debate over the price discovery process between equity and CDS markets, we detect credit-driven price discovery in equity markets. We demonstrate that price discovery process is more credit market driven when a company’s credit risk increases, which is significantly more prominent for small-sized firms with highly volatile equity price, and increasing default probability. At last, we turn to study news impact on the trend and the volatility in commodity market. Applying an innovated textual machine learning to business news articles related to corn markets, we extract topics from news. We demonstrate that textual news about financial markets, soybean-biofuel, crop progress and exports significantly contributes in explaining the corn price dynamics. Our volatility analysis demonstrates that soybean and biofuel media coverage contributes also positively to the level of uncertainty regarding corn price. We conclude that news items related to this topic generally provide outlook information leaving scope for interpretation and thus uncertainty after their release.