Herding is a feature of investor behavior in financial markets, particularly in market stress. In the thesis we will apply an approach based on the transversal dispersion of individual stock betas, which allows us to extract herd patterns, using two dynamic methodologies to measure the herd phenomenon over time with a state-space model in three different markets. The first study analyzes beta herding in the Brazilian stock market using a statespace model, controlled by two groupings of companies: those stocks listed on the market index (Ibovespa) and those listed on the stock exchange as a whole. The findings revealed a high herd on the Brazilian stock exchange, with only small differences between the clusters. Regarding the control variables, we found that the dividend yield, market volatility, SMB, and WML factors were significant for both groups, indicating that the herd is significant regardless of the behavior of these variables. The second study examines beta herding in the commodity market, using the methodology developed by Hwang and Salmon (2004) and a beta adaptation standardized by Hwang, Rubesam, and Salmon (2018). We analyzed the behavior of fifteen commodities between 2000 and 2018 and then extracted the food commodities to test their effect separately. The results suggest that betas can deviate from fundamentals in both samples. However, food commodity betas tend to revert more quickly to stability between demand and supply, which results in a long-term risk-return balance. The third study applies the methodology of Hwang and Salmon (2004) and a beta adaptation standardized by Hwang, Rubesam, and Salmon (2018) for the Cryptocurrency market. The results reveal that the herd towards the market presents significant movement and persistence regardless of the market condition, expressed through the market index, market volatility, and volatility index. When analyzing trail herding, it is possible to observe that herding was intense during the investigated period. We also identified a positive relationship between herding and market stress.