scholarly journals Profit uplift modeling for direct marketing campaigns: approaches and applications for online shops

2021 ◽  
Author(s):  
Daniel Baier ◽  
Björn Stöcker
Keyword(s):  
Negative Binomial ◽  
Direct Marketing ◽  
Sample Selection ◽  
Negative Binomial Regression ◽  
Uplift Modeling ◽  
Binomial Regression ◽  
Ordinary Regression ◽  
Real World Datasets ◽  
Traditional Approaches ◽  
Online Shops

AbstractIn order to select “best” customers for a direct marketing campaign, response models are widespread: a sample of customers receives an ad, a catalog, a sample pack, or a discount offer on a test basis. Then, their responses (e.g., website visits, conversions, or revenues) are used to build a predictive model. Finally, this model is applied to all customers in order to select “best” ones for the campaign. However, up to now, only models that reflect website visits, conversions, or revenues have been proposed. In this paper, we discuss the shortcomings of these traditional approaches and propose profit uplift modeling appoaches based on one-stage ordinary regression and random forests as well as two-stage Heckman sample selection and zero-inflated negative binomial regression for parameter estimation. The new approaches demonstrate superiority to the traditional ones when applied to real-world datasets. One dataset reflects recent discount offers of a large online fashion retailer. The other is the well-known Hillstrom dataset that describes two Email campaigns.

scholarly journals Is area deprivation associated with greater impacts of COVID-19 in care homes across England? A preliminary analysis of COVID-19 outbreaks and deaths

2021 ◽  
pp. jech-2020-215039 ◽  
Author(s):  
Anders Malthe Bach-Mortensen ◽  
Michelle Degli Esposti
Keyword(s):  
Negative Binomial ◽  
Care Home ◽  
Negative Binomial Regression ◽  
Population Characteristics ◽  
Care Homes ◽  
Future Research ◽  
Area Deprivation ◽  
Binomial Regression ◽  
Index Of Multiple Deprivation

IntroductionThe COVID-19 pandemic has disproportionately impacted care homes and vulnerable populations, exacerbating existing health inequalities. However, the role of area deprivation in shaping the impacts of COVID-19 in care homes is poorly understood. We examine whether area deprivation is linked to higher rates of COVID-19 outbreaks and deaths among care home residents across upper tier local authorities in England (n=149).MethodsWe constructed a novel dataset from publicly available data. Using negative binomial regression models, we analysed the associations between area deprivation (Income Deprivation Affecting Older People Index (IDAOPI) and Index of Multiple Deprivation (IMD) extent) as the exposure and COVID-19 outbreaks, COVID-19-related deaths and all-cause deaths among care home residents as three separate outcomes—adjusting for population characteristics (size, age composition, ethnicity).ResultsCOVID-19 outbreaks in care homes did not vary by area deprivation. However, COVID-19-related deaths were more common in the most deprived quartiles of IDAOPI (incidence rate ratio (IRR): 1.23, 95% CI 1.04 to 1.47) and IMD extent (IRR: 1.16, 95% CI 1.00 to 1.34), compared with the least deprived quartiles.DiscussionThese findings suggest that area deprivation is a key risk factor in COVID-19 deaths among care home residents. Future research should look to replicate these results when more complete data become available.

scholarly journals Association between fatality rate of COVID-19 and selenium deficiency in China

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Hai-Yang Zhang ◽  
An-Ran Zhang ◽  
Qing-Bin Lu ◽  
Xiao-Ai Zhang ◽  
Zhi-Jie Zhang ◽  
...  
Keyword(s):  
Negative Binomial ◽  
Viral Infections ◽  
Case Fatality ◽  
Selenium Deficiency ◽  
Negative Binomial Regression ◽  
Negative Binomial Regression Model ◽  
Fatality Rate ◽  
Selenium Content ◽  
Individual Level ◽  
Binomial Regression

Abstract Background COVID-19 has impacted populations around the world, with the fatality rate varying dramatically across countries. Selenium, as one of the important micronutrients implicated in viral infections, was suggested to play roles. Methods An ecological study was performed to assess the association between the COVID-19 related fatality and the selenium content both from crops and topsoil, in China. Results Totally, 14,045 COVID-19 cases were reported from 147 cities during 8 December 2019–13 December 2020 were included. Based on selenium content in crops, the case fatality rates (CFRs) gradually increased from 1.17% in non-selenium-deficient areas, to 1.28% in moderate-selenium-deficient areas, and further to 3.16% in severe-selenium-deficient areas (P = 0.002). Based on selenium content in topsoil, the CFRs gradually increased from 0.76% in non-selenium-deficient areas, to 1.70% in moderate-selenium-deficient areas, and further to 1.85% in severe-selenium-deficient areas (P < 0.001). The zero-inflated negative binomial regression model showed a significantly higher fatality risk in cities with severe-selenium-deficient selenium content in crops than non-selenium-deficient cities, with incidence rate ratio (IRR) of 3.88 (95% CIs: 1.21–12.52), which was further confirmed by regression fitting the association between CFR of COVID-19 and selenium content in topsoil, with the IRR of 2.38 (95% CIs: 1.14–4.98) for moderate-selenium-deficient cities and 3.06 (1.49–6.27) for severe-selenium-deficient cities. Conclusions Regional selenium deficiency might be related to an increased CFR of COVID-19. Future studies are needed to explore the associations between selenium status and disease outcome at individual-level.

scholarly journals Statistical models for analyzing count data: predictors of length of stay among HIV patients in Portugal using a multilevel model

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Ahmed Nabil Shaaban ◽  
Bárbara Peleteiro ◽  
Maria Rosario O. Martins
Keyword(s):  
Length Of Stay ◽  
Regression Model ◽  
Random Effects ◽  
Count Data ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Comprehensive Approach ◽  
Negative Binomial Regression Model ◽  
Hiv Patients ◽  
Binomial Regression

Abstract Background This study offers a comprehensive approach to precisely analyze the complexly distributed length of stay among HIV admissions in Portugal. Objective To provide an illustration of statistical techniques for analysing count data using longitudinal predictors of length of stay among HIV hospitalizations in Portugal. Method Registered discharges in the Portuguese National Health Service (NHS) facilities Between January 2009 and December 2017, a total of 26,505 classified under Major Diagnostic Category (MDC) created for patients with HIV infection, with HIV/AIDS as a main or secondary cause of admission, were used to predict length of stay among HIV hospitalizations in Portugal. Several strategies were applied to select the best count fit model that includes the Poisson regression model, zero-inflated Poisson, the negative binomial regression model, and zero-inflated negative binomial regression model. A random hospital effects term has been incorporated into the negative binomial model to examine the dependence between observations within the same hospital. A multivariable analysis has been performed to assess the effect of covariates on length of stay. Results The median length of stay in our study was 11 days (interquartile range: 6–22). Statistical comparisons among the count models revealed that the random-effects negative binomial models provided the best fit with observed data. Admissions among males or admissions associated with TB infection, pneumocystis, cytomegalovirus, candidiasis, toxoplasmosis, or mycobacterium disease exhibit a highly significant increase in length of stay. Perfect trends were observed in which a higher number of diagnoses or procedures lead to significantly higher length of stay. The random-effects term included in our model and refers to unexplained factors specific to each hospital revealed obvious differences in quality among the hospitals included in our study. Conclusions This study provides a comprehensive approach to address unique problems associated with the prediction of length of stay among HIV patients in Portugal.

scholarly journals Association between breakfast skipping and metabolic outcomes by sex, age, and work status stratification

2021 ◽  
Vol 18 (1) ◽  
Author(s):  
Jun Heo ◽  
Won-Jun Choi ◽  
Seunghon Ham ◽  
Seong-Kyu Kang ◽  
Wanhyung Lee
Keyword(s):  
Regression Analysis ◽  
Negative Binomial ◽  
Metabolic Diseases ◽  
Young Male ◽  
Negative Binomial Regression ◽  
Final Analysis ◽  
Work Status ◽  
Breakfast Skipping ◽  
Metabolic Outcomes ◽  
Binomial Regression

Abstract Background The association between breakfast skipping and abnormal metabolic outcomes remains controversial. A comprehensive study with various stratified data is required. Objective The aim of this study was to investigate the relationship between abnormal metabolic outcomes and breakfast skipping by sex, age, and work status stratification. Methods We used data from the Korea National Health and Nutrition Examination Surveys from 2013 to 2018. A total of 21,193 (9022 men and 12,171 women) participants were included in the final analysis. The risk of metabolic outcomes linked to breakfast skipping was estimated using the negative binomial regression analysis by sex, work status, and age stratification. Results A total of 11,952 (56.4%) participants consumed breakfast regularly. The prevalence of abnormal metabolic outcomes was higher among those with irregular breakfast consumption habits. Among young male workers, negative binomial regression analysis showed that irregular breakfast eaters had a higher risk of abnormal metabolic outcomes, after adjusting for covariates (odds ratio, 1.15; 95% confidence interval, 1.03–1.27). Conclusions The risk of abnormal metabolic outcomes was significant in young men in the working population. Further studies are required to understand the association of specific working conditions (working hours or shift work) with breakfast intake status and the risk of metabolic diseases.

scholarly journals Coronary deaths during Midsummer festival in Finland: miseries of long, light nights

2021 ◽  
Author(s):  
Simo Näyhä
Keyword(s):  
Coronary Heart Disease ◽  
Heart Disease ◽  
Alcohol Consumption ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Adverse Consequence ◽  
Heavy Alcohol ◽  
Summer Peak ◽  
Binomial Regression ◽  
Heavy Alcohol Consumption

AbstractThis paper examines whether the anomalous summer peak in deaths from coronary heart disease (CHD) in Finland could be attributed to adverse effects of the Midsummer festival and alcohol consumption during the festival. Daily deaths from CHD and alcohol poisoning in Finland, 1961–2014, that occurred during the 7 days centering on Midsummer Day were analysed in relation to deaths during 14 to 4 days before and 4 to 14 after Midsummer Day. Daily counts of deaths from CHD among persons aged 35–64 years were regressed on days around the Midsummer period by negative binomial regression. Mortality from CHD was highest on Midsummer Day (RR 1.25 (95% confidence interval 1.12–1.31), one day after the peak in deaths from alcohol poisonings. RR for CHD on Midsummer Day was particulary high (RR = 1.43; 1.09–1.86) in the 2000s, 30% of deaths being attributable to that day. In conclusion, the anomalous and prominent summer peak in deaths from CHD in Finland is an adverse consequence of the Midsummer festival. The most likely underlying reason is heavy alcohol consumption during the festival period, especially on Midsummer Eve. In the 2000s, one third of deaths from CHD on Midsummer Day are preventable.

scholarly journals Mobility and social deprivation on primary care utilisation among paediatric patients with asthma

2021 ◽  
Vol 9 (3) ◽  
pp. e001085
Author(s):  
Jennifer A Lucas ◽  
Miguel Marino ◽  
Sophia Giebultowicz ◽  
Katie Fankhauser ◽  
Shakira F Suglia ◽  
...  
Keyword(s):  
Primary Care ◽  
Influenza Vaccination ◽  
Negative Binomial ◽  
Social Deprivation ◽  
Pneumococcal Vaccination ◽  
Negative Binomial Regression ◽  
High Deprivation ◽  
Care Utilisation ◽  
Binomial Regression ◽  
Children With Asthma

ObjectiveAsthma care is negatively impacted by neighbourhood social and environmental factors, and moving is associated with undesirable asthma outcomes. However, little is known about how movement into and living in areas of high deprivation relate to primary care use. We examined associations between neighbourhood characteristics, mobility and primary care utilisation of children with asthma to explore the relevance of these social factors in a primary care setting.DesignIn this cohort study, we conducted negative binomial regression to examine the rates of primary care visits and annual influenza vaccination and logistic regression to study receipt of pneumococcal vaccination. All models were adjusted for patient-level covariates.SettingWe used data from community health centres in 15 OCHIN states.ParticipantsThe sample included 23 773 children with asthma aged 3–17 across neighbourhoods with different levels of social deprivation from 2012 to 2017. We conducted negative binomial regression to examine the rates of primary care visits and annual influenza vaccination and logistic regression to study receipt of pneumococcal vaccination. All models were adjusted for patient-level covariates.ResultsClinic visit rates were higher among children living in or moving to areas with higher deprivation than those living in areas with low deprivation (rate ratio (RR) 1.09, 95% CI 1.02 to 1.17; RR 1.05, 95% CI 1.00 to 1.11). Children moving across neighbourhoods with similarly high levels of deprivation had increased RRs of influenza vaccination (RR 1.13, 95% CI 1.03 to 1.23) than those who moved but stayed in neighbourhoods of low deprivation.ConclusionsMovement into and living within areas of high deprivation is associated with more primary care use, and presumably greater opportunity to reduce undesirable asthma outcomes. These results highlight the need to attend to patient movement in primary care visits, and increase neighbourhood-targeted population management to improve equity and care for children with asthma.

Evaluation of Strategies to Mitigate Culvert-Involved Crashes

2021 ◽  
pp. 036119812199207
Author(s):  
Hitesh Chawla ◽  
Megat-Usamah Megat-Johari ◽  
Peter T. Savolainen ◽  
Christopher M. Day
Keyword(s):  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Cost Effective ◽  
Negative Binomial Regression Model ◽  
Promising Alternative ◽  
Roadside Safety ◽  
Adverse Conditions ◽  
Traffic Characteristics ◽  
Annual Average Daily Traffic ◽  
Binomial Regression

The objectives of this study were to assess the in-service safety performance of roadside culverts and evaluate the potential impacts of installing various safety treatments to mitigate the severity of culvert-involved crashes. Such crashes were identified using standard fields on police crash report forms, as well as through a review of pertinent keywords from the narrative section of these forms. These crashes were then linked to the nearest cross-drainage culvert, which was associated with the nearest road segment. A negative binomial regression model was then estimated to discern how the risk of culvert-involved crashes varied as a function of annual average daily traffic, speed limit, number of travel lanes, and culvert size and offset. The second stage of the analysis involved the use of the Roadside Safety Analysis Program to estimate the expected crash costs associated with various design contexts. A series of scenarios were evaluated, culminating in guidance as to the most cost-effective treatments for different combinations of roadway geometric and traffic characteristics. The results of this study provide an empirical model that can be used to predict the risk of culvert-involved crashes under various scenarios. The findings also suggest that the installation of safety grates on culvert openings provides a promising alternative for most of the cases where the culvert is located within the clear zone. In general, a guardrail is recommended when adverse conditions are present or when other treatments are not feasible at a specific location.

Systemic Determinants in Surgery: Nonclinical Factors Affecting Time to Operation for Incarcerated Ventral Hernias

2021 ◽  
pp. 000313482110111
Author(s):  
David E. Wang ◽  
Paul J. Chung ◽  
Rafael Barrera ◽  
Gene F. Coppa ◽  
Antonio E. Alfonso ◽  
...  
Keyword(s):  
Operating Room ◽  
Negative Binomial ◽  
Private Insurance ◽  
Negative Binomial Regression ◽  
Outcome Variable ◽  
Hospital Size ◽  
Factors Affecting ◽  
Binomial Regression ◽  
Comorbidity Status ◽  
Ventral Hernias

Introduction We explore nonclinical factors affecting the amount of time from admission to the operating room for patients requiring nonelective repair of ventral hernias. Methods Using the 2005-2012 Nationwide Inpatient Sample, we identified adult patients with a primary diagnosis of ventral hernia without obstruction/gangrene, who underwent nonelective repair. The outcome variable of interest was time from admission to surgery. We performed univariate and multivariable analyses using negative binomial regression, adjusting for age, sex, race, income, insurance, admission day, comorbidity status (van Walraven score), diagnosis, procedure, hospital size, location/teaching status, and region. Results 7,253 patients met criteria, of which majority were women (n = 4,615) and white (n = 5,394). The majority of patients had private insurance (n = 3,015) followed by Medicare (n = 2,737). Median time to operation was 0 days. Univariate analysis comparing operation <1 day to ≥1 day identified significant differences in race, day of admission, insurance, length of stay, comorbidity status, hospital location, type, and size. Negative binomial regression showed that weekday admission (IRR 4.42, P < .0001), private insurance (IRR 1.53-2.66, P < .0001), rural location (IRR 1.39-1.76, P < .01), small hospital size (IRR 1.26-1.36, P < .05), white race (IRR 1.30-1.34, P < .01), healthier patients (van Walraven score IRR 1.05, P < .0001), and use of mesh (IRR 0.39-0.56, P < .02) were associated with shorter time until procedure. Conclusion Shorter time from admission to the operating room was associated with several nonclinical factors, which suggest disparities may exist. Further prospective studies are warranted to elucidate these disparities affecting patient care.

Sign in / Sign up

Export Citation Format

Share Document