Major public health emergencies always test the credibility of the government. The success of governments’ strategies relies on trust in government and broad acceptance of response measures. The profound experience of the epidemic often has a long-term impact on people’s cognition. We construct a difference-in-difference estimator by combining the variations of epidemic effects across cohorts and regions, and intend to evaluate the long-term effect of individuals’ early SARS experience on trust in government during the COVID-19 pandemic. We also use the instrumental variable method to overcome the endogenous problem caused by two-way causality. The results show that the impact of COVID-19 has significantly reduced trust in government of the groups who had not been exposed to the SARS epidemic (including groups who were in early childhood and the unborn during the SARS outbreak). While it has a positive impact on trust in government of people experienced SARS in adolescence, and only a little negative impact on trust in government of people experienced SARS in adulthood. We also find that the impact of COVID-19 mainly reduced the trust in government among groups socially vulnerable or without SARS experience (e.g., low income, low social status etc.). The results suggest that: (a) the trust created by governments’ successful anti-epidemic measures is long-lasting; (b) governments should pay more attention to their trust among socially vulnerable groups.
We consider a financial market with memory effects in which wealth processes are driven by mean-field stochastic Volterra equations. In this financial market, the classical dynamic programming method can not be used to study the optimal investment problem, because the solution of mean-field stochastic Volterra equation is not a Markov process. In this paper, a new method through Malliavin calculus introduced in [1], can be used to obtain the optimal investment in a Volterra type financial market. We show a sufficient and necessary condition for the optimal investment in this financial market with memory by mean-field stochastic maximum principle.
Many statistical problems can be formulated as the missing data problems. The data augmentation algorithm and the inverse Bayes formula are important tools for constructing iterative optimization or samplings via the introduction of unobserved data or latent variables. As a Markov Chain Monte Carlo method, the data augmentation algorithm has its autocorrelation. Nevertheless, the convergence rate of the data augmentation algorithm is not known in the Genetic Linkage Model, and the same is true for the inverse Bayes formula method. In this article, we analyze the convergence rates of the data augmentation algorithm and the inverse Bayes formula method in the genetic linkage model, and through simulation results we get their convergence rates.