In order to fully capture the degree of risk in the financial market, we use the asymmetric Laplace distribution (ALD) to describe the distribution characteristics of the financial rate of return in combination with a generalized autoregressive score (GAS) model. We apply the GAS-ALD model to time-varying-parameter rolling estimation of the risk measures value-at-risk (VaR) and expected shortfall (ES), with the aim of building a dynamic risk measurement model for the financial market during the Covid-19 pandemic. Data for representative industry indexes of the Shanghai Stock Exchange are selected for empirical analysis, and parametric and nonparametric methods are used for comparison and backtesting. The results show, first, that the GAS-ALD model has obvious advantages over other tested methods in estimating VaR and ES and predicting the volatility risk of the rate of return, with ES being more accurate at predicting risk in extreme cases than VaR. Second, the pandemic had a big impact on the raw materials and energy industries, but a smaller impact on the financial industry. Based on the research conclusions, recommendations are put forward for three areas: government, regulators and enterprises.