The COVID-19 pandemic has had a long-term impact on industries worldwide,
with the hospitality and food industry facing significant challenges, leading
to the permanent closure of many restaurants and the loss of jobs. In this
study, we developed an innovative analytical framework using Hamiltonian Monte
Carlo for predictive modeling with Bayesian regression, aiming to estimate the
change point in consumer behavior towards different types of restaurants due to
COVID-19. Our approach emphasizes a novel method in computational analysis,
providing insights into customer behavior changes before and after the
pandemic. This research contributes to understanding the effects of COVID-19 on
the restaurant industry and is valuable for restaurant owners and policymakers.

本研究使用 Hamiltonian Monte Carlo 和贝叶斯回归构建创新的分析框架，旨在估计 COVID-19 对不同类型餐厅顾客行为的变化点，从而为餐厅业主和政策制定者提供对疫情影响的理解。

餐饮行业中 COVID-19 对消费行为影响的贝叶斯回归方法

A Bayesian Regression Approach for Estimating the Impact of COVID-19 on  Consumer Behavior in the Restaurant Industry

The digital revolution has led to the digitization of human behavior,
creating unprecedented opportunities to understand observable actions on an
unmatched scale. Emerging phenomena such as crowdfunding and crowdsourcing have
further illuminated consumer behavior while also introducing new behavioral
patterns. However, the sheer volume and complexity of this data present
significant challenges for marketing researchers and practitioners. Traditional
methods used to analyze consumer data fall short in handling the breadth,
precision, and scale of emerging data sources. To address this, computational
methods have been developed to manage the "big data" associated with consumer
behavior, which typically includes structured data, textual data, audial data,
and visual data. These methods, particularly machine learning, allow for
effective parsing and processing of multi-faceted data. Given these recent
developments, this review article seeks to familiarize researchers and
practitioners with new data sources and analysis techniques for studying
consumer behavior at scale. It serves as an introduction to the application of
computational social science in understanding and leveraging publicly available
consumer data.

数字革命已经导致了人类行为的数字化，出现了众筹和众包等新增的行，这给营销研究和实践者带来了前所未有的机会，但同时也面临着数据过于庞杂和复杂的挑战。为了解决这些问题，开发了计算方法以管理与消费者行为相关的 “大数据” 等技术，特别是机器学习，使得对多方面数据的有效解析和处理成为可能。本文综述了新的数据来源和分析技术，旨在将计算社会科学应用于理解和利用公开的消费者数据。

机器学习与消费者数据

Machine Learning and Consumer Data

Consumers interact with firms across multiple devices, browsers, and
machines; these interactions are often recorded with different identifiers for
the same consumer. The failure to correctly match different identities leads to
a fragmented view of exposures and behaviors. This paper studies the identity
fragmentation bias, referring to the estimation bias resulted from using
fragmented data. Using a formal framework, we decompose the contributing
factors of the estimation bias caused by data fragmentation and discuss the
direction of bias. Contrary to conventional wisdom, this bias cannot be signed
or bounded under standard assumptions. Instead, upward biases and sign
reversals can occur even in experimental settings. We then compare several
corrective measures, and discuss their respective advantages and caveats.

研究了消费者行为在多设备、浏览器和机器上交互时所面临的身份识别偏差，解构了数据碎片化造成的估计偏差因素并讨论了其方向，提出了几项纠正措施及其各自的优缺点。