Predicting Brain Drain in Investment Professionals: A Machine Learning based Decision Making Using the Theory of Planned Behaviour

Abstract

"The phenomenon of brain drain has garnered increasing attention due to its significant socio-economic implications, and especially in Sri Lanka due to the large plethora of migration the country has witnessed over the past few years exacerbated due to the crisis faced by the country. This thesis delves into the predictive modeling of brain drain among investment professionals in Sri Lanka, utilizing the Theory of Planned Behavior (TPB) as a conceptual framework. The study aims to bridge the gap in existing literature by focusing on predictive modeling techniques, which have been relatively underexplored in the context of brain drain, particularly in Sri Lanka's financial services industry.

Through an extensive literature review, this research identifies the push-pull and endogenous-exogenous factors influencing the decision-making processes of investment professionals regarding migration. The TPB is adopted as a theoretical lens to understand the complex interplay of attitudes, social norms, and perceived behavioral control in shaping migration intentions.

The primary research question revolves around the effectiveness of a classification-based machine learning approach in modeling the decision-making processes of investment professionals and predicting brain drain. The study's objectives include investigating the factors contributing to brain drain in the financial services industry, evaluating the TPB's applicability in predicting migration intentions, and developing a machine learning-based prediction model.

Data collection involved a survey of investment professionals in Sri Lanka, with machine learning techniques such as Logistic Regression, Gradient Boosting, Random Forest, Decision Trees, Naïve Bayes, and Support Vector Machine employed to develop predictive models – with Gradient Boosting algorithm being chosen as the best performing algorithm. The study found that certain factors, including general attitude towards migration, access to better living standards, parental opinions on migration, and personal beliefs regarding resources, significantly influence migration intentions.

Furthermore, the research highlights the importance of addressing brain drain from both organizational and policy perspectives. Employers can implement strategies such as improving remuneration and providing international secondment opportunities to retain valuable talent. Meanwhile, policymakers should focus on enhancing the country's economic and political climate, infrastructure, and quality of life to mitigate brain drain. In conclusion, this thesis contributes to the understanding of brain drain among investment professionals in Sri Lanka and provides valuable insights for employers, policymakers, and future researchers aiming to address this critical issue."