Report submission

Introduction

SDG No Poverty: Goal 1 – Eradicate Poverty

The aim was to assess the ability of families to bear expenses, taking into account the total value of their salaries and expenses. 

Indicator Selected: Positions held by women in senior management positions with two sub indicators Board Members and Executives
Objective: To reduce inequalities within and between countries.
Indicator (y): Total poverty intensity rate.

Selected variables (x):
1) Poverty risk rate among the employed population. This variable measures the proportion of employed individuals who are still at risk of poverty.
2) Rate of severe material and social deprivation. This variable measures the proportion of the population experiencing severe material and social deprivation, such as lack of access to necessities or social exclusion.
3) Gender pay gap. This variable measures the difference in average earnings between men and women.

Data Preprocessing

Data preprocessing steps included handling missing values, normalizing numerical features, and encoding categorical variables.  

Machine Learning Techniques

It was used all the mentioned models, as asked:

1. Decision Tree
2. K-Nearest Neighbors (KNN)
3. Neural Networks

Machine Learning Models

1.Decision Tree

·         Accuracy: low.

2. K-Nearest Neighbors (KNN)

·         Accuracy: medium.

 3.Neural Networks

In this case, a neural network was trained up to 90 iterations, achieving a value close to 484,620 (at least with a tendency to reach it as the iterations progress). However, it is observed that, as the training progresses from 40 iterations onwards, its improvement is not significant. This indicates that training the data cannot improve its cost.

• Accuracy: very low.

Observations

The variables have an impact on the "Poverty Intensity Rate," but it cannot be said whether this impact is negative or positive. However, given the nature of the objective itself, it can be assumed that when one of the Xs (1 to 3) tends towards poor values (increase in poverty risk rate or decrease in minimum wage, or no increase in it), this also negatively impacts the Poverty Intensity Rate (causing it to increase).

(I'm not sure that the Neural Networks were quite accurate when calculated.)

In an effort to understand the factors contributing to poverty and inequality, this study delved into the data pertaining to Sustainable Development Goal 1: "End poverty in all its forms everywhere." The analysis focused on three key indicators: poverty risk rate among the employed population, rate of severe material and social deprivation, and gender pay gap. These variables were selected to capture different dimensions of poverty and inequality, providing a comprehensive overview of the issue. The study highlighted the potential influence of additional variables on the observed outcomes, emphasizing the interconnectedness of factors that contribute to poverty and inequality. The relationship between these variables was explored to gain insights into the underlying causes of the observed patterns.

Particularly noteworthy is the connection between poverty and household financial ratios, as well as the monthly financial balance of families. The study suggests that a country's poverty rate may be directly linked to the average household financial ratio. This implies that the difference between financial inflows and outflows plays a significant role in determining the level of poverty within a population. By examining these relationships, the study aims to contribute to a deeper understanding of the complex factors that contribute to poverty and inequality, paving the way for more effective and targeted interventions to address these challenges.