-
Compositional dynamic modelling for causal prediction in multivariate time series
Authors:
Kevin Li,
Graham Tierney,
Christoph Hellmayr,
Mike West
Abstract:
Theoretical developments in sequential Bayesian analysis of multivariate dynamic models underlie new methodology for causal prediction. This extends the utility of existing models with computationally efficient methodology, enabling routine exploration of Bayesian counterfactual analyses with multiple selected time series as synthetic controls. Methodological contributions also define the concept…
▽ More
Theoretical developments in sequential Bayesian analysis of multivariate dynamic models underlie new methodology for causal prediction. This extends the utility of existing models with computationally efficient methodology, enabling routine exploration of Bayesian counterfactual analyses with multiple selected time series as synthetic controls. Methodological contributions also define the concept of outcome adaptive modelling to monitor and inferentially respond to changes in experimental time series following interventions designed to explore causal effects. The benefits of sequential analyses with time-varying parameter models for causal investigations are inherited in this broader setting. A case study in commercial causal analysis-- involving retail revenue outcomes related to marketing interventions-- highlights the methodological advances.
△ Less
Submitted 4 June, 2024;
originally announced June 2024.
-
Reply to Discussions of "Multivariate Dynamic Modeling for Bayesian Forecasting of Business Revenue"
Authors:
Anna K. Yanchenko,
Graham Tierney,
Joseph Lawson,
Christoph Hellmayr,
Andrew Cron,
Mike West
Abstract:
We are most grateful to all discussants for their positive comments and many thought-provoking questions. In addition, the discussants provide a number of useful leads into various areas of the literatures on time series, forecasting and commercial application within which the work in our paper is, of course, just one contribution linked to multiple threads. Our view is that, collectively, the dis…
▽ More
We are most grateful to all discussants for their positive comments and many thought-provoking questions. In addition, the discussants provide a number of useful leads into various areas of the literatures on time series, forecasting and commercial application within which the work in our paper is, of course, just one contribution linked to multiple threads. Our view is that, collectively, the discussion contributions nicely expand on the core of the paper and together -- with multiple additional references -- provide an excellent point-of-entrée to the broader field of retail forecasting and its research challenges. Interested readers are encouraged to dig deeply into the discussions and our responses here, and explore referenced sources.
There are several themes that recur across discussants, as well as a range of specific points/questions raised. Following some "big-picture" comments on our perspectives on Bayesian forecasting systems, we comment in turn on some specifics in each contribution.
△ Less
Submitted 2 April, 2023;
originally announced April 2023.
-
Multivariate Bayesian dynamic modeling for causal prediction
Authors:
Graham Tierney,
Christoph Hellmayr,
Greg Barkimer,
Kevin Li,
Mike West
Abstract:
Bayesian forecasting is developed in multivariate time series analysis for causal inference. Causal evaluation of sequentially observed time series data from control and treated units focuses on the impacts of interventions using contemporaneous outcomes in control units. Methodological developments here concern multivariate dynamic models for time-varying effects across multiple treated units wit…
▽ More
Bayesian forecasting is developed in multivariate time series analysis for causal inference. Causal evaluation of sequentially observed time series data from control and treated units focuses on the impacts of interventions using contemporaneous outcomes in control units. Methodological developments here concern multivariate dynamic models for time-varying effects across multiple treated units with explicit foci on sequential learning and aggregation of intervention effects. Analysis explores dimension reduction across multiple synthetic counterfactual predictors. Computational advances leverage fully conjugate models for efficient sequential learning and inference, including cross-unit correlations and their time variation. This allows full uncertainty quantification on model hyper-parameters via Bayesian model averaging. A detailed case study evaluates interventions in a supermarket promotions experiment, with coupled predictive analyses in selected regions of a large-scale commercial system. Comparisons with existing methods highlight the issues of appropriate uncertainty quantification in casual inference in aggregation across treated units, among other practical concerns.
△ Less
Submitted 19 June, 2024; v1 submitted 6 February, 2023;
originally announced February 2023.
-
Multivariate Dynamic Modeling for Bayesian Forecasting of Business Revenue
Authors:
Anna K. Yanchenko,
Graham Tierney,
Joseph Lawson,
Christoph Hellmayr,
Andrew Cron,
Mike West
Abstract:
Forecasting enterprise-wide revenue is critical to many companies and presents several challenges and opportunities for significant business impact. This case study is based on model developments to address these challenges for forecasting in a large-scale retail company. Focused on multivariate revenue forecasting across collections of supermarkets and product Categories, hierarchical dynamic mod…
▽ More
Forecasting enterprise-wide revenue is critical to many companies and presents several challenges and opportunities for significant business impact. This case study is based on model developments to address these challenges for forecasting in a large-scale retail company. Focused on multivariate revenue forecasting across collections of supermarkets and product Categories, hierarchical dynamic models are natural: these are able to couple revenue streams in an integrated forecasting model, while allowing conditional decoupling to enable relevant and sensitive analysis together with scalable computation. Structured models exploit multi-scale modeling to cascade information on price and promotion activities as predictors relevant across Categories and groups of stores. With a context-relevant focus on forecasting revenue 12 weeks ahead, the study highlights product Categories that benefit from multi-scale information, defines insights into when, how and why multivariate models improve forecast accuracy, and shows how cross-Category dependencies can relate to promotion decisions in one Category impacting others. Bayesian modeling developments underlying the case study are accessible in custom code for interested readers.
△ Less
Submitted 10 December, 2021;
originally announced December 2021.
-
Reconstructing evolving signalling networks by hidden Markov nested effects models
Authors:
Xin Wang,
Ke Yuan,
Christoph Hellmayr,
Wei Liu,
Florian Markowetz
Abstract:
Inferring time-varying networks is important to understand the development and evolution of interactions over time. However, the vast majority of currently used models assume direct measurements of node states, which are often difficult to obtain, especially in fields like cell biology, where perturbation experiments often only provide indirect information of network structure. Here we propose hid…
▽ More
Inferring time-varying networks is important to understand the development and evolution of interactions over time. However, the vast majority of currently used models assume direct measurements of node states, which are often difficult to obtain, especially in fields like cell biology, where perturbation experiments often only provide indirect information of network structure. Here we propose hidden Markov nested effects models (HM-NEMs) to model the evolving network by a Markov chain on a state space of signalling networks, which are derived from nested effects models (NEMs) of indirect perturbation data. To infer the hidden network evolution and unknown parameter, a Gibbs sampler is developed, in which sampling network structure is facilitated by a novel structural Metropolis--Hastings algorithm. We demonstrate the potential of HM-NEMs by simulations on synthetic time-series perturbation data. We also show the applicability of HM-NEMs in two real biological case studies, in one capturing dynamic crosstalk during the progression of neutrophil polarisation, and in the other inferring an evolving network underlying early differentiation of mouse embryonic stem cells.
△ Less
Submitted 5 May, 2014;
originally announced May 2014.
