Economists in different fields have been developing theoretical models and conducting experimental studies to understand how people are making decisions over various products and different domains. The first essay explores how individual choices may change over time and differ across multiple domains. This study also shows evidence for domain-specific temporal discounting and suggests that consumers are generally more impatient for health rewards but more patient for environmental rewards. The second essay aims to explore which models, the traditional temporal discounting models, or machine learning models, can better predict individuals’ intertemporal choices. Results suggests that some machine learning algorithms, for example, random forest, have better prediction powers compared to the temporal discounting models. Although machine learning method sometimes suffer from overfitting problems, it may have the potential to give more accurate predictions for individual choices when the training data have enough information on individual previous choices or behaviors. The third essay examines how information framing and consumers’ neighborhood attachment impact their product choices. In particular, the product we focus on is low-input turfgrass. We find the presence of environmental impact information in the choice experiment has impacts on homeowners’ preferences for low-input turfgrass. We also find the homeowners’ neighborhood attachment affects their lawn maintenance behavior.