The Economics of Psychological Well-Being Evidence from the United States Using Machine Learning Method
This paper models peoples psychological well-being (PWB) by applying machine learning methods to a large dataset in the United States. PWB can be quantified by three self-reported variables: general happiness, satisfaction with the financial situation, and satisfaction with the job. First, we run the K-nearest neighbor (KNN) algorithm for each respective feature to rank the importance of features. We found that marital status has a higher importance score than income for peoples general happiness. Prestige score of occupation is the most important predictor of satisfaction with jobs. Next, we utilize the Forward Selection algorithm to find the best combination of predictions. Using this selected combination to predict people’s PWB, we achieve 70% - 80% classification accuracy when detecting people with disadvantaged PWB. On top of that, we use ordered probit regression to quantify how each feature a↵ects PWB. Lastly, we show that PWB is important to economics by investigating how happiness affects physical and mental health, risky goods consumption, investment decisions, and working behaviors. We find that happier people have better health conditions, smoke and drink less, have more confidence in financial institutions, and generally work more hours
Year of publication: |
[2022]
|
---|---|
Authors: | Yang, Yanchao |
Publisher: |
[S.l.] : SSRN |
Subject: | Künstliche Intelligenz | Artificial intelligence | USA | United States | Wirtschaftspsychologie | Economic psychology |
Saved in:
freely available
Saved in favorites
Similar items by subject
-
Innovation in the Public Sector : Smarter States, Services and Citizens
Demir, Fatih, (2022)
-
Administration by Algorithm? Public Management Meets Public Sector Machine Learning
Veale, Michael, (2019)
-
Essays on applied production analysis
Ley, Eduardo, (1991)
- More ...