Grocery Data Research Articles

Using Grocery Data for Credit Decisions

Many consumers across the world struggle to gain access to credit because of their lack of credit scores. This paper explores the potential of a new alternative data source, grocery transaction data, for evaluating consumers’ creditworthiness. Our analysis takes advantage of a unique, individual-level match of credit card data and supermarket loyalty card data. By developing credit scoring algorithms that either exclude or include grocery data, we illustrate both the incremental value of grocery data for credit decisions and its boundary conditions. We demonstrate that signals from grocery data can improve credit approval decisions, particularly for individuals who lack traditional credit scores. Furthermore, as a consumer establishes a relationship with lenders and builds a credit history, the marginal value of incorporating grocery data diminishes. These findings highlight the potential of grocery data in informing credit decisions and, consequently, in enabling financial institutions to extend credit to consumers who lack traditional credit scores. This paper was accepted by David Simchi-Levi, marketing. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.02364 .

The effect of cross‐border shopping on commodity tax revenue: results from Norway's COVID‐19 border closings

AbstractWe use grocery data from Norway and COVID‐19 border closings to gauge the effect of cross‐border shopping on commodity tax revenue. Detailed store–category‐level data identify differential treatment effects that depend on distance to Swedish stores. Economically significant effects extend to up to two hours' drive from the border, and even further for prominent cross‐border shopping products, such as beer, cigarettes, and carbonated soft drinks. Across all products, cross‐border shopping decreases tax revenue from VAT by 3.6 percent at the national level. National commodity tax revenue from carbonated soft drinks (subject to a sugar tax) is reduced by 8.1 percent and from cigarettes by 11.9 percent.

Transient Customer Response to Data Breaches of Their Information

Cybersecurity breaches pose a substantial concern in the digital era. We investigate how customers respond to multiple unexpected data breaches of their information in India. Difference-in-differences estimates show that digital payments declined by 9% relative to cash payments immediately after an unexpected data breach in a food delivery platform, but the gap disappeared three months later. Customer entry and exit also exhibit weak, short-lived changes. Additional analyses on bank and online grocery data breaches uncover even weaker effects of data breaches. Our findings imply that the perceived benefits of convenience outweigh the costs of payment security risks. This paper was accepted by David Sraer, finance. Funding: This work was supported by the MOE (Ministry of Education in China) Project of Humanities and Social Sciences [Project 20YJC790183], National University of Singapore [Start-Up Grant A-0003870-00-00], the National Natural Science Foundation of China [Grant 72150004], and the National Nature Science Foundation of China [Grant 72203112]. Supplemental Material: The online appendices and data are available at https://doi.org/10.1287/mnsc.2021.01335 .

Optimizing an mHealth Intervention to Change Food Purchasing Behaviors for Cancer Prevention: Protocol for a Pilot Randomized Controlled Trial.

BackgroundDietary intake is a powerful modifiable factor that influences cancer risk; however, most US adults do not adhere to dietary guidelines for cancer prevention. One promising pathway for improving dietary adherence is targeting grocery shopping habits. Interventions might facilitate healthy grocery choices, with a combination of mHealth and traditional methods, by promoting the salience of dietary goals while shopping, enhancing motivation to make dietary changes, and increasing household support for healthy food purchasing.ObjectiveThis pilot study will assess feasibility and acceptability of intervention components designed to improve adherence to dietary guidelines for cancer prevention (preliminary aim). The primary aim of the study is to quantify the effect of each intervention component, individually and in combination, on dietary intake (primary aim) and grocery store food purchases (exploratory aim). Mediation analyses will be conducted to understand the mechanisms of action (goal salience, motivation, and household support—secondary aims). The overarching goal is to optimize an mHealth intervention to be tested in a future fully powered clinical trial.MethodsThe study enrolled adults (N=62) with low adherence to dietary recommendations for cancer prevention. In a 20-week program, all participants attend a nutrition education workshop and receive weekly educational messages through an app. A factorial design is used to test 4 intervention components: (1) location-triggered messages: educational messages are delivered when arriving at grocery stores; (2) reflections on the benefits of change: content is added to messages to encourage reflection on anticipated benefits of healthy eating, and participants attend an additional workshop session and 3 coach calls on this topic; (3) coach monitoring: food purchases are monitored digitally by a coach who sends personalized weekly app messages and conducts 3 coaching calls that focus on feedback about purchases; and (4) household support: another adult in the household receives messages designed to elicit support for healthy food purchasing, and support is addressed in 3 coach calls and an extra workshop session attended by the index participant and household member. Assessments are completed at weeks 0, 10, and 20 using self-report measures, as well as objective capture of grocery data from the point of purchase using store loyalty accounts.ResultsThe National Cancer Institute funded this study (R21CA252933) on July 7, 2020. Participant recruitment began in the spring of 2021 and concluded with the successful enrollment of 62 participants. Data collection is expected to be completed in the summer of 2022, and results are expected to be disseminated in the summer of 2023.ConclusionsThe results of this study will inform the development of scalable interventions to lower cancer risk via changes in dietary intake.Trial RegistrationClinicalTrials.gov NCT04947150; https://clinicaltrials.gov/ct2/show/NCT04947150International Registered Report Identifier (IRRID)DERR1-10.2196/39669

Buying and Payment Habits: Using Grocery Data to Predict Credit Card Payments

This study shows that individuals' habits in grocery shopping are incrementally useful in predicting their credit card payment behaviors and that such incremental predictive power can translate into incremental profits for firms. Guided by prior work, we identify five broad grocery shopping habits that are correlated with payment behaviors: (1) shopping the same day of week, (2) relying on a shopping budget, (3) consistently buying the same brands and categories, (4) taking advantage of deals and promotions, and (5) buying healthier products. Knowledge of the five grocery habits offers guidance on how to transform the raw grocery data into inputs for flexible machine learning models, which we use to assess the incremental predictive power of grocery data. We find the incremental predictive gain from grocery data, above and beyond standard data sets used by issuers, ranges from 0.2% to 9.4%, depending on the data environment faced by issuers in various credit markets. Furthermore, simulations of issuers' credit extension decisions illustrate that the marginal impact on issuer profits ranges from 0.3% to 15.2% and is greatest for consumers who do not have an established credit history. This suggests that grocery data may enable credit card issuers to extend credit to consumers who currently have limited or no access to credit. We also discuss a boundary condition in which grocery data may not have incremental value. Overall, this study highlights how consumer data from a seemingly unrelated domain can help address a managerial problem in the focal domain.

A study on forecasting bigmart sales using optimized machine learning techniques

Data mining is an in-depth study of enormous amounts of data present in an organization or institution’s repository. Business experts mostly utilize data analytics approaches to confirm their opinion. It will rapidly boost the global interest of the organization. In this scenario, the information and conclusion are gathered from Data analysis by data analytics. The experts also use it to validate, diagnose, or authenticate speculate layouts suppositions and completion of the analysis. In this paper, the prediction is based on grocery data sets by inspecting and analyzing the big mart sales data set. Among several predictive algorithms, data mining algorithms are considered for prediction. It includes Decision Tree, Naive Bayes, Adaboost with Particle Swarm optimization, and Random forest. The proposed method of this research is a novel Naive Bayes with a PSO algorithm. This algorithm optimizes the model iteratively. Exploration of the data must be done before prediction. The root means squared error (RMSE) is used as evaluation metrics for comparing the data mining algorithms. The proposed algorithm performs well and gives a lower RMSE value. So, the proposed algorithm fits the best model when compared with the existing algorithms. This paper describes the prediction of high-quality data analysis data and determines the efficiency of data mining algorithms.

A Decade of U.S. Natural Disasters and Household Food-at-Home Expenditures and Quality: A Quasi-Experimental Study

This study exploits spatial and temporal variation in natural disasters in the United States via a generalized differences-in-differences approach to identify the impact of natural disasters on households’ food-at-home (FAH) spending and quality from 2005 to 2016. We use two datasets: (i) the Storm Events Database to identify U.S. counties that experience severe economic losses as a result of droughts, floods, hurricanes, and tornadoes, and (ii) the Nielsen Consumer Panel Data for grocery data. We find that only floods and hurricanes affect FAH spending. Floods (Hurricanes) have a persistent (immediate) effect on FAH spending. On average, highly damaging floods (hurricanes) decrease 15-day FAH spending by about $2 ($7) in 90 days (30 days) after the events. The FAH quality effect of the four natural disasters is either inconsequential or nonexistent. We provide indirect evidence that the FAH spending effect of natural disasters works through both income and price channels. We also find that hurricanes have an anticipation effect on total grocery spending which starts 15 days before the disaster date. Our results are robust to an alternative specification that controls for county-specific linear trends. We add to the growing body of literature on the effects of natural disasters on household finances and financial decisions. Our findings could be of particular interest to post-disaster relief organizations and their programs.

Privacy versus Convenience: Customer Response to Data Breaches of Their Information

Cybersecurity breaches pose a substantial privacy concern in the digital era. We investigate how customers respond to privacy leakage in multiple unexpected data breaches. Difference-in-differences estimates show that digital payments declined by 4.6% ~ 7.5% relative to cash payments immediately after an unexpected data breach in a food delivery platform, but the gap disappeared three months later. Customer entry and exit also exhibit weak, short-lived changes. Additional analyses on bank and online grocery data breaches uncover even weaker direct and spillover effects of data breaches. Our findings imply that the perceived benefit of convenience overweighs the cost of privacy leakages.

A meta-analysis of correlations between market share and other brand performance metrics in FMCG markets

The strong relationship between market share and other metrics (i.e., penetration rate, purchase frequency, share of requirements) is well-documented in the marketing literature. However, there is no systematic meta-analysis of this important empirical generalization. This study quantifies the category-level correlations between market share and the aforementioned brand metrics. The category-level effects of promotional intensity, existence of sub-segments, purchase volume variation, and competition from private label and niche brands on these correlations are tested using grocery data from more than 400 categories. These factors have a negative effect all in three observed correlations with one exception. Promotional activity increases the correlation between market share and share of requirements. These results further generalize our knowledge of “Double Jeopardy” in FMCG markets and support the advice to brand managers to concentrate on increasing penetration rather than purchase frequency or share of requirements.

Cohort Analysis: Analyzing Data On Product Use

Palmores cohort analysis technique is a practitioner-oriented method to unravel age, period, and cohort effects in the data on product use. The technique is fully explained, and towards the end of the article, it is applied to grocery data for illustration purposes.

Finding Associations in Composite Data Sets

In this paper, a composite fuzzy association rule mining mechanism (CFARM), directed at identifying patterns in datasets comprised of composite attributes, is described. Composite attributes are defined as attributes that can take simultaneously two or more values that subscribe to a common schema. The objective is to generate fuzzy association rules using “properties” associated with these composite attributes. The exemplar application is the analysis of the nutrients contained in items found in grocery data sets. The paper commences with a review of the back ground and related work, and a formal definition of the CFARM concepts. The CFARM algorithm is then fully described and evaluated using both real and synthetic data sets.

The feasibility and utility of grocery receipt analyses for dietary assessment

ObjectiveTo establish the feasibility and utility of a simple data collection methodology for dietary assessment.DesignUsing a cross-sectional design, trained data collectors approached adults (~20 – 40 years of age) at local grocery stores and asked whether they would volunteer their grocery receipts and answer a few questions for a small stipend ($1).MethodsThe grocery data were divided into 3 categories: "fats, oils, and sweets," "processed foods," and "low-fat/low-calorie substitutions" as a percentage of the total food purchase price. The questions assessed the shopper's general eating habits (eg, fast-food consumption) and a few demographic characteristics and health aspects (eg, perception of body size).Statistical Analyses Performed. Descriptive and analytic analyses using non-parametric tests were conducted in SAS.ResultsForty-eight receipts and questionnaires were collected. Nearly every respondent reported eating fast food at least once per month; 27% ate out once or twice a day. Frequency of fast-food consumption was positively related to perceived body size of the respondent (p = 0.02). Overall, 30% of the food purchase price was for fats, oils, sweets, 10% was for processed foods, and almost 6% was for low-fat/low-calorie substitutions. Households where no one was perceived to be overweight spent a smaller proportion of their food budget on fats, oils, and sweets than did households where at least one person was perceived to be overweight (p = 0.10); household where the spouse was not perceived to be overweight spent less on fats, oils, and sweets (p = 0.02) and more on low-fat/low-calorie substitutions (p = 0.09) than did households where the spouse was perceived to be overweight; and, respondents who perceived themselves to be overweight spent more on processed foods than did respondents who did not perceive themselves to be overweight (p = 0.06).ConclusionThis simple dietary assessment method, although global in nature, may be a useful indicator of dietary practices as evidenced by its association with perceived weight status.

Early statistical detection of anthrax outbreaks by tracking over-the-counter medication sales.

The recent series of anthrax attacks has reinforced the importance of biosurveillance systems for the timely detection of epidemics. This paper describes a statistical framework for monitoring grocery data to detect a large-scale but localized bioterrorism attack. Our system illustrates the potential of data sources that may be more timely than traditional medical and public health data. The system includes several layers, each customized to grocery data and tuned to finding footprints of an epidemic. We also propose an evaluation methodology that is suitable in the absence of data on large-scale bioterrorist attacks and disease outbreaks.

Habit Formation and Dynamic Demand Functions

Most economists would agree that past consumption patterns are an important determinant of present consumption patterns, and that one ought to distinguish between long-run and short-run demand functions. But although the distinction between long-run and short-run behavior is traditional in the theory of the firm, it is seldom made in the theory of consumer behavior. If we regard demand theory as a theory of how a given amount of money (expenditure, called income) is allocated among goods, then-in a world without consumer durables-there are three reasons why longand shortrun demand functions might differ. (i) The consumer may have contractually fixed commitments which prevent him from adjusting some portion of his consumption (for example, housing) in response to changes in prices or income. When these fixed commitments lapse, he is able to adjust to his long-run equilibrium. (ii) The consumer may be ignorant of consumption possibilities or of his own tastes outside the range of his past consumption experience. In this case his adjustment to a new price-income situation will involve a time-consuming learning process. (iii) Finally, goods may be "habit forming" so that an individual's current preferences depend on his past consumption patterns. In this case a change in prices or income will cause a change-in consumption which will induce a change in tastes, which will cause a further change in consumption. In this paper I formulate a model of consumer behavior based on habit formation, beginning with a specific class of demand functions derived from the "modified Bergson family" of utility functions. The properties of these utility functions and the corresponding demand functions are briefly summarized in Section 1. I then postulate that the parameters