Forecasting Key Retail Performance Indicators Using Interpretable Regression.

Foot traffic, conversion rate, and total sales during a period of time may be considered to be important indicators of store performance. Forecasting them may allow for business managers plan stores operation in the near future in an efficient way. This work presents a regression method that is able to predict these three indicators based on previous data. The previous data includes values for the indicators in the recent past; therefore, it is a requirement to have gathered them in a suitable manner. The previous data also considers other values that are easily obtained, such as the day of the week and hour of the day of the indicators. The novelty of the approach that is presented here is that it provides a confidence interval for the predicted information and the importance of each parameter for the predicted output values, without additional processing or analysis. Real data gathered by Follow Up, a customer experience company, was used to test the proposed method. The method was tried for making predictions for up to one month in the future. The results of the experiments show that the proposed method has a comparable performance to the best methods proposed in the past that do not provide confidence intervals or parameter rankings. The method obtains RMSE of 0.0713 for foot traffic prediction, 0.0795 for conversion rate forecasting, and 0.0757 for sales prediction.

Air pollution and hospital emergency room and admissions for cardiovascular and respiratory diseases in Doña Ana County, New Mexico.

INTRODUCTION: Doña Ana County in New Mexico regularly experiences severe air pollution episodes associated with windblown dust and fires. Residents of Hispanic/Latino origin constitute the largest population group in the region. We investigated the associations of ambient particulate matter and ozone with hospital emergency room and admissions for respiratory and cardiovascular visits in adults. METHODS: We used trajectories regression analysis to determine the local and regional components of particle mass and ozone. We applied Poisson generalized models to analyze hospital emergency room visits and admissions adjusted for pollutant levels, humidity, temperature and temporal and seasonal effects. RESULTS: We found that the sources within 500km of the study area accounted for most of particle mass and ozone concentrations. Sources in Southeast Texas, Baja California and Southwest US were the most important regional contributors. Increases of cardiovascular emergency room visits were estimated for PM10 (3.1% (95% CI: -0.5 to 6.8)) and PM10-2.5 (2.8% (95% CI: -0.2 to 5.9)) for all adults during the warm period (April-September). When high PM10 (>150µg/m(3)) mass concentrations were excluded, strong effects for respiratory emergency room visits for both PM10 (3.2% (95% CI: 0.5-6.0)) and PM2.5 (5.2% (95% CI: -0.5 to 11.3)) were computed. CONCLUSIONS: Our analysis indicated effects of PM10, PM2.5 and O3 on emergency room visits during the April-September period in a region impacted by windblown dust and wildfires.