The objective of this project is to develop machine learning models to predict the prevalence of mental disorders in the population based on economic and demographic factors such as GDP, unemployment rate, country population, and year. Additionally, it aims to predict the consumption of anxiolytics in Spain in the near future.
Different datasets from various sources have been used, including the OECD to gather economic and demographic information from several countries, and the WHO to obtain information on the prevalence of mental disorders in each country. Furthermore, a dataset from the Spanish Ministry of Health has been used, which collects information on anxiolytic consumption monthly from 2010 to 2021.
Two machine learning models have been developed: a Random Forest model to predict the prevalence of mental disorders in the population based on the mentioned economic and demographic factors, and another model based on an Autoregressive Integrated Moving Average (ARIMA) model to predict the demand for anxiolytics in Spain.
The models have been evaluated using different metrics, including Mean Absolute Error (MAE) and Mean Squared Error (MSE), and their ability to make accurate predictions has been confirmed.
This project has potential applications in the field of public health, as it can help identify the factors that contribute to the prevalence of mental disorders and the consumption of anxiolytics in the population. Additionally, the developed models can be used by governments and health organizations to plan and develop more effective health policies.