Guizhi Wang

With the acceleration of urbanization, air pollution, especially PM_2.5, has seriously affected human health and reduced people's life quality. Accurate PM_2.5 prediction is significant for environmental protection authorities to take actions and develop prevention countermeasures. In this article, an adapted Kalman filter (KF) approach is presented to remove the nonlinearity and stochastic uncertainty of time series, suffered by the autoregressive integrated moving average (ARIMA) model. To further improve the accuracy of PM_2.5 forecasting, a hybrid model is proposed by introducing an autoregressive (AR) model, where the AR part is used to determine the state-space equation, whereas the KF part is used for state estimation on PM_2.5 concentration series. A modified artificial neural network (ANN), called AR-ANN is introduced to compare with the AR-KF model. According to the results, the AR-KF model outperforms the AR-ANN model and the original ARIMA model on the predication accuracy; that is, the AR-ANN obtains 10.85 and 15.45 of mean absolute error and root mean square error, respectively, whereas the ARIMA gains 30.58 and 29.39 on the corresponding metrics. It, therefore, proves that the presented AR-KF model can be adopted for air pollutant concentration prediction.

Since 2013, the concentration of PM_2.5 in Beijing has been has dropped significantly, there is still there is still far from the World Health Organization’s revised 2021 target of 5µg/m 3 of the average annual concentration. We analyzed the monthly, quarterly and annual trends of PM_2.5 concentrations in the Beijing, China, from 2014 to 2021 using independent T-test. In addition, we proposed a new combined PM_2.5 concentration prediction model, which combines mode decomposition and reconstruction algorithm, convolutional neural network and long short-term memory neural networks. The results show that: (1) From 2016 to 2018, PM_2.5 concentration in Beijing decreased at a fast pace, and then the decline entered a slow period. (2) The combination of the modal decomposition algorithm and neural network model can effectively process the PM_2.5 concentration data with the characteristics of nonlinearity, instability and high complexity, and is more stable and generalizable. (3) The combination of convolutional neu-ral network and long short-term memory neural network model has a higher prediction accuracy than traditional neural network models. The research results are helpful to know the possible air pollution process in advance and take corresponding measures to reduce the harm caused by air pollution in the new stage of haze pollution.

In this study, we propose a new gray model called “normalized beta function gray model”. Its main idea is to use the normalized beta function to optimize the linear gray action quantity into a nonlinear form. It is worth mentioning that when solving the nonlinear parameters of the newly proposed model, we construct a nonlinear optimization problem with constraints by taking the mean square error of the model as the objective function, and find the optimal solution under the given precision by using the simulated annealing algorithm. We use the new gray model to predict gasoline consumption and PM2.5 concentration, compared with the other three gray models, the proposed gray model reduces the mean absolute percentage error to 2.99% and 32.75%, respectively, which is 88.67% and 60.25% lower than the maximum value of mean absolute percentage error of four gray models. We also apply the newly proposed gray model to the simulation studies, which shows excellent prediction ability.

The random forest (RF) algorithm is a very practical and excellent ensemble learning algorithm. In this paper, we improve the random forest algorithm and propose an algorithm called ‘post-selection boosting random forest’ (PBRF). This algorithm combines the original random forest and the Lasso method, without giving the number of decision trees for final prediction in advance, it can dynamically obtain the decision trees according to different input samples to output the prediction results. Meanwhile, we verify that the proposed algorithm can improve the performance of the model through simulation studies and real data analysis.

At present particulate matter (PM_2.5) pollution represents a serious threat to the public health and the national economic system in China. This paper optimizes the whitening coefficient in a grey Markov model by a genetic algorithm, predicts the concentration of fine particulate matter (PM_2.5), and then quantifies the health effects of PM_2.5 pollution by utilizing the predicted concentration, computable general equilibrium (CGE), and a carefully designed exposure–response model. Further, the authors establish a social accounting matrix (SAM), calibrate the parameter values in the CGE model, and construct a recursive dynamic CGE model under closed economy conditions to assess the long-term economic losses incurred by PM_2.5 pollution. Subsequently, an empirical analysis was conducted for the Beijing area: Despite the reduced concentration trend, PM_2.5 pollution continued to cause serious damage to human health and the economic system from 2013 to 2020, as illustrated by various facts, including: (1) the estimated premature deaths and individuals suffering haze pollution-related diseases are 156,588 (95% confidence intervals (CI): 43,335–248,914)) and six million, respectively; and (2) the accumulated labor loss and the medical expenditure negatively impact the regional gross domestic product, with an estimated loss of 3062.63 (95% CI: 1,168.77–4671.13) million RMB. These findings can provide useful information for governmental agencies to formulate relevant environmental policies and for communities to promote prevention and rescue strategies.

Econometrics and input–output models have been presented to construct a joint model (i.e., an EC + IO model) in the paper, which is characterized by incorporating the uncertainty of the real economy with the detailed departmental classification structure, as well as adding recovery period variables in the joint model to make the model dynamic. By designing and implementing a static model, it is estimated that the indirect economic loss for the transportation sector caused by representative haze pollution of Beijing in 2013 was 23.7 million yuan. The industrial-related indirect losses due to the direct economic losses incurred by haze pollution reached 102 million yuan. With the constructed dynamic model, the cumulative economic losses for the industrial sectors have been calculated for the recovery periods of different durations. The results show that: (1) the longer the period that an industrial department returns to normal output after haze pollution has impacted, the greater the cumulative economic loss will be; (2) when the recovery period is one year, the cumulative economic loss value computed by the dynamic EC + IO model is much smaller than the loss value obtained by the static EC + IO model; (3) the recovery curves of industrial sectors show that the recovery rate at the early stage is fast, while it is slow afterwards. Therefore, the governance work after the occurrence of haze pollution should be launched as soon as possible. This study provides a theoretical basis for evaluating the indirect economic losses of haze pollution and demonstrates the value of popularization and application.

In this paper, a vector autoregression (VAR) model has been constructed in order to analyze a two-way mechanism between PM_2.5 pollution and industry development in Beijing via the combination of an impulse response function and variance decomposition. According to the results, long-term equilibrium interconnection was found between PM_2.5 pollution and the development of primary, secondary, and tertiary industries. One-way Granger causalities were found in the three types of industries shown to contribute to PM_2.5 pollution, though the three industries showed different scales of influences on the PM_2.5 pollution that varied for about 1–2 years. The development of the primary and secondary industries increased the emission of PM_2.5, but the tertiary industry had an inhibitory effect. In addition, PM_2.5 pollution had a certain inhibitory effect on the development of the primary and secondary industries, but the inhibition of the tertiary industry was not significant. Therefore, the development of the tertiary industry can contribute the most to the reduction of PM_2.5 pollution. Based on these findings, policy-making recommendations can be proposed regarding upcoming pollution prevention strategies.