IIM-Shillong

In the aftermath of the global financial crisis in 2008, a great deal of research has been conducted to assess the financial vulnerability of households across the world. Given the multidimensional nature of vulnerability, researchers have focused on different dimensions and approaches to analyse the status of vulnerability of the households.

The proposed work attempts to analyse household’s financial vulnerability employing the machine learning and deep learning techniques using the data from a financial services provider which covers household-level information about demographics, income, expenditure, savings and borrowings for its clients for the states of Tamil Nadu, Odisha, and Uttarakhand.

The vulnerability analysis is carried out in two parts. In the first part, homogenous clusters of households are identified and analysed to further identify the cluster of financially vulnerable households. In the second part, the households’ probability of being financially vulnerable based on economic and demographic characteristics is estimated.

The results of the proposed research would be important for designing and implementation of financial sector policies focused on redistribution of income and strengthening the risk management strategies of the households.