Time-series characteristics and optimal investment strategies for livestock stocks

This study focused on the optimal strategy of livestock stock portfolios, where several representative livestock stocks were selected to construct portfolios and their time-series characteristics were explored through time-series analysis methods, including spectral analysis to identify periodic and non-periodic characteristics in stock returns. Based on modern portfolio theory, we proposed a novel utility function that integrates conditional value at risk (CVaR), risk aversion, and loss sensitivity into the classical utility model. Combined with the historical return data, we estimated the specific expressions of each index about the weight vector, and calculated the optimal weight allocation of the portfolio in different periods through the gradient descent method for investors with different degrees of risk aversion and loss sensitivity. The results show that non-periodic stocks have a relative advantage in terms of expected return and volatility control, and as investors' risk aversion increases, the investment weights are gradually tilted toward low-volatility stocks. This study provides empirical support for the investment strategy of livestock stocks, and provides a reference for investors to achieve a balance between risk control and return optimization in a complex and volatile market environment.