Predicting Suicide Incidence in the Philippines Using Random Forest Algorithm

Abstract

This study determined the trend of suicide in the Philippines and identified which national indicators are possible predictors of suicide incidences. The indicators considered were a mix of objective and subjective indicators. The objective indicators consisted of Real GDP Per Capita, Unemployment Rate, and Volume of Crime while the subjective ones were Perceived Political Instability, Self-Rated Poverty, and Net Personal Optimism Scores. Data were drawn from World Bank Open Data, and records of Philippine Statistics Authority, Philippine National Police, and Social Weather Station. From 2006 to 2021, it was found that the number of suicide deaths more than doubled. The most significant upturn in deaths was observed during the first year of the pandemic, 2020. In addition, the random forest regression model found for the data, it appeared that unemployment, political instability, net personal optimism, and real GDP per capita can predict suicide deaths. Consistent with Durkheim’s theory on suicide, the decline in economic well-being of people and an increase in their perceived political instability were found to be related to an increase in suicide deaths. Furthermore, in line with the Hopelessness Theory of Suicide, people’s low optimism towards the future quality of their life was also found to be related to increased suicide deaths. Thus, these indicators should be monitored, and relevant government bodies should employ strategies and programs to raise per capita income, create more jobs, make more people employable, manage perceptions of political stability, and make people more optimistic about their lives.