We are sometimes forced to use the Interrupted Time Series (ITS) design as an identification strategy for potential policy change, such as when we only have a single treated unit and no comparable controls. For example, with recent county- and state-wide criminal justice reform efforts, where judicial bodies have changed bail setting practices for everyone in their jurisdiction in order to reduce rates of pre-trial detention while maintaining court order and public safety, we have no natural comparison group other than the past. In these contexts, it is imperative to model pre-policy trends with a light touch, allowing for structures such as autoregressive departures from any pre-existing trend, in order to accurately and realistically assess the statistical uncertainty of our projections (beyond the stringent assumptions necessary for the subsequent causal inferences). To tackle this problem we provide a methodological approach rooted in commonly understood and used modeling approaches that better captures uncertainty. We quantify uncertainty with simulation, generating a distribution of plausible counterfactual trajectories to compare to the observed; this approach naturally allows for incorporating seasonality and other time varying covariates, and provides confidence intervals along with point estimates for the potential impacts of policy change. We find simulation provides a natural framework to capture and show uncertainty in the ITS designs. It also allows for easy extensions such as nonparametric smoothing in order to handle multiple post-policy time points or more structural models to account for seasonality.
A marginalized three-part interrupted time series regression model for proportional data
Oregon Health & Science University(俄勒冈健康与科学大学);Kaiser Permanente Washington Health Research Institute(凯撒永久华盛顿健康研究所);University of Washington(华盛顿大学);King’s College London(伦敦国王学院)
Bracketing in the Comparative Interrupted Time-Series Design to Address Concerns about History Interacting with Group: Evaluating Missouri Handgun Purchaser Law
Multiple-group (Controlled) Interrupted Time Series Analysis with Higher-Order Autoregressive Errors: A Simulation Study Comparing Newey-West and Prais-Winsten Methods
University of California, San Francisco(加州大学旧金山分校)
Multilevel non-linear interrupted time series analysis
Washington University in St. Louis School of Medicine(华盛顿大学圣路易斯医学院);Center for Advancing Health Services, Policy & Economics Research, Washington University in St. Louis School of Medicine(华盛顿大学圣路易斯医学院健康服务、政策与经济学研究中心);Cardiovascular Division, Department of Medicine, Washington University in St. Louis School of Medicine(华盛顿大学圣路易斯医学院医学部心血管科);School of Public Health, Washington University in St. Louis(华盛顿大学公共卫生学院);Brown School of Social Work, Washington University in St. Louis(华盛顿大学布朗社会工作学院)
Post-Model-Selection Statistical Inference with Interrupted Time Series Designs: An Evaluation of an Assault Weapons Ban in California
媒体上有很多关于枪支销售变化原因的说法,也有一些值得尊敬的研究。因果推断的挑战可能相当艰巨。本文使用中断的时间序列设计和分析报告了1996年至2018年加利福尼亚州每日手枪销售的分析。该设计于1963年由Campbell和Stanley介绍给社会科学家,分析方法于1975年由Box和Tiao提出,并且很容易找到更近的治疗方法(Box et al.,2016)。但这种因果推断的方法可能会被手枪销售数据严重击败,特别是当因果影响被估计时。对本文来说,更重要的是所采用的标准统计方法中的基本疏忽。测试多重性问题是由推荐实践中的自适应模型选择引入的。挑战是计算和概念。在这两个问题上都取得了一些进展,可以说是对过去研究的改进,但带回家的信息可能是减少对可以学到什么的期望。
Statistical Power for Estimating Treatment Effects Using Difference-in-Differences and Comparative Interrupted Time Series Designs with Variation in Treatment Timing
Choosing an analytic approach: Key study design considerations in state policy evaluation
Rutgers Institute for Health, Health Care Policy and Aging Research(罗格斯大学健康、医疗保健政策与老龄化研究所);Department of Psychiatry, Rutgers Robert Wood Johnson Medical School(罗格斯鲁布-怀特医学精神病学系);RAND Corporation(兰德公司);Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health(约翰霍普金斯大学彭博公共卫生学院生物统计学系)
A Bayesian changepoint methodology for high dimensional multivariate time series and space-time data: A study of structural change using remotely sensed data