Time Series Treatment Effects Analysis with Always-Missing Controls
This work addresses a specific challenge in time series causal inference for applications like retail analytics, offering a method applicable to always-missing control scenarios and other settings.
The paper tackles the problem of estimating treatment effects in time series data when control groups are always unobservable, such as during holidays, by recovering the control group in event periods while handling confounders and temporal dependencies. Results on M5 Walmart retail sales data show robust estimation of potential outcomes and accurate holiday effect predictions, with theoretical guarantees for consistency and asymptotic normality.
Estimating treatment effects in time series data presents a significant challenge, especially when the control group is always unobservable. For example, in analyzing the effects of Christmas on retail sales, we lack direct observation of what would have occurred in late December without the Christmas impact. To address this, we try to recover the control group in the event period while accounting for confounders and temporal dependencies. Experimental results on the M5 Walmart retail sales data demonstrate robust estimation of the potential outcome of the control group as well as accurate predicted holiday effect. Furthermore, we provided theoretical guarantees for the estimated treatment effect, proving its consistency and asymptotic normality. The proposed methodology is applicable not only to this always-missing control scenario but also in other conventional time series causal inference settings.