Cover cropping has the potential to improve resilience of agriculture to climate‐change‐induced extreme weather events. However, rigorous quantitative evidence on the resilience effect of cover crops is still lacking. Using a novel data set that combines satellite‐based cover crop information and county‐level crop insurance data, we examine the impact of planting cover crops on prevented‐planting‐related losses that are typically caused by heavy rainfall events. The US federal crop insurance program offers “prevented planting” coverage, which pays indemnities if insured growers are unable to plant their crop due to adverse weather. Linear fixed effects models, instrument‐based estimation methods, long‐difference models, and a number of other robustness checks are utilized in the empirical analysis to achieve the study objective. Our findings suggest that counties with higher cover crop adoption rates tend to have lower levels of crop insurance losses due to prevented planting. The resulting reduction in prevented planting risk also becomes larger with longer term, multiyear cover crop use. These results support the notion that cover crops improve soil conditions such that the likelihood and magnitude of prevented planting losses decrease. We posit that the ability of cover crops to handle excess moisture (i.e., through better water absorption and improved water infiltration in the soil) is the main factor in its ability to reduce prevented planting losses in the US Midwest.

Cerebrovascular accidents (CVA) cause a range of impairments in coordination, such as a spectrum of walking impairments ranging from mild gait imbalance to complete loss of mobility. Patients with CVA need personalized approaches tailored to their degree of walking impairment for effective rehabilitation. This paper aims to evaluate the validity of using various machine learning (ML) and deep learning (DL) classification models (support vector machine, Decision Tree, Perceptron, Light Gradient Boosting Machine, AutoGluon, SuperTML, and TabNet) for automated classification of walking assistant devices for CVA patients. We reviewed a total of 383 CVA patients’ (1623 observations) prescription data for eight different walking assistant devices from five hospitals. Among the classification models, the advanced tree-based classification models (LightGBM and tree models in AutoGluon) achieved classification results of over 90% accuracy, recall, precision, and F1-score. In particular, AutoGluon not only presented the highest predictive performance (almost 92% in accuracy, recall, precision, and F1-score, and 86.8% in balanced accuracy) but also demonstrated that the classification performances of the tree-based models were higher than that of the other models on its leaderboard. Therefore, we believe that tree-based classification models have potential as practical diagnosis tools for medical rehabilitation.

No abstract

No abstract