Multiple psychological characteristics predict housing mortgage loan behavior: A holistic model based on machine learning.

The factors that influence consumers' house choice are debatable. Previous studies have examined the effects of demographic and socioeconomic attributes, physical and environmental features of the house, and isolated single psychological characteristics on housing behavior. However, these factors are still not sufficient to predict consumer housing behavior, particularly when they are measured separately. We construct a holistic model that integrates psychological characteristics including values, personality traits, motivation, decision-making style, and risk-seeking together with demographic and socioeconomic factors to jointly predict housing mortgage loan behavior. This study aims to use a newly developed statistical method, "machine learning," to examine the relationship between multiple psychological characteristics and consumer housing mortgage loan behavior. Data were collected through an online survey (N = 2,270). The results show that the holistic psychological model is effective for predicting consumer housing mortgage loan behavior in the life context. Moreover, by analyzing and comparing the relative impact of all predictors, we find that psychological characteristics made a more important contribution to predicting housing mortgage loan behavior than did traditional factors (demographic and socioeconomic factors). The results provide a new perspective for understanding the effects of how multiple psychological characteristics integrally predict consumers' housing mortgage loan behavior in the real estate market. Theoretical and practical implications for marketing and sales are discussed.