Air pollution has emerged as one of the major public health threats. In recent years, Bangladesh has ranked among top air polluted countries around the globe. Air quality in Bangladesh is measured on daily basis using national standards of five pollutants—NO 2 , CO, Ozone (O 3 ), SO 2 , and particulate matter (PM 2.5 and PM 10 ) and presented as indexed value ranged from 0 to 500 which is further classified from good to extremely unhealthy. In this study, we have applied Seasonal Autoregressive Integrated Moving Average (SARIMA) model to forecast weekly air quality of Dhaka and Sylhet divisions in Bangladesh. Our study reveals the existence of seasonal pattern of the air quality and suggests extremely unhealthy and very unhealthy air in January–March 2020. Therefore, policymaker should address this period as the air quality directly influences public health.

中文翻译：

---

预测孟加拉国达卡和锡尔赫特地区的空气质量：时间序列建模方法

空气污染已成为主要的公共卫生威胁之一。近年来，孟加拉国已跻身全球空气污染最严重的国家之列。孟加拉国的空气质量每天使用五种污染物——NO 2 、CO、臭氧 (O 3 )、SO 2 和颗粒物（PM 2.5 和 PM 10 ）的国家标准进行测量，并以指数值表示，范围从 0 到 500这进一步从良好到极不健康分类。在这项研究中，我们应用季节性自回归综合移动平均 (SARIMA) 模型来预测孟加拉国达卡和锡尔赫特地区的每周空气质量。我们的研究揭示了空气质量季节性模式的存在，并表明 2020 年 1 月至 3 月的空气非常不健康和非常不健康。因此，