Wang Joanna J J, Fung Thomas, Weatherburn Donald
School of Mathematical Physical Sciences, University of Technology, Sydney, Australia.
Department of Mathematics and Statistics, Macquarie University, Sydney, NSW Australia.
Crime Sci. 2021;10(1):24. doi: 10.1186/s40163-021-00160-x. Epub 2021 Oct 24.
The spread of COVID-19 has prompted Governments around the world to impose draconian restrictions on business activity, public transport, and public freedom of movement. The effect of these restrictions appears to vary from country to country and, in some cases, from one area to another within a country. This paper examines the impact of the COVID-19 restrictions imposed in New South Wales (NSW) by the State Government. We examine week-to-week changes in 13 categories of crime (and four aggregated categories) from 2 January 2017 to 28 June 2020. Rather than using the pre-intervention data to make a forecast and then comparing that with what is actually observed, we use a Box-Jenkins (ARIMA) approach to model the entire time series. Our results are broadly in accord with those of other studies, but we find no effect of the lockdown (upward or downward) on domestic assault.
The online version contains supplementary material available at 10.1186/s40163-021-00160-x.
新冠疫情的蔓延促使世界各国政府对商业活动、公共交通和公众行动自由实施严厉限制。这些限制的效果在不同国家似乎有所不同，在某些情况下，在一个国家内的不同地区也存在差异。本文研究了新南威尔士州（NSW）州政府实施的新冠疫情限制措施的影响。我们考察了2017年1月2日至2020年6月28日期间13类犯罪（以及四类汇总犯罪）的周度变化。我们不是用干预前的数据进行预测，然后将其与实际观察结果进行比较，而是采用Box-Jenkins（自回归积分移动平均）方法对整个时间序列进行建模。我们的结果与其他研究大致一致，但我们发现封锁措施对家庭暴力没有影响（无论是上升还是下降）。
在线版本包含可在10.1186/s40163-021-00160-x获取的补充材料。
