The dataset comprises estimates of household, food service, and retail waste per country, extracted from the 2021 UNEP Food Waste Index Report. It encompasses information on waste generation, confidence levels, and economic indicators such as GDP per capita, lending rates, and more.
- Handling Missing Values:
- Identified and addressed missing values through normalization or removal.
- Ensuring Data Consistency:
- Verified and standardized data types and formats for consistency.
- Addressing Outliers:
- Explored and mitigated outliers that could impact analysis results.
This dataset serves as a valuable resource for studying the relationship between economic indicators and waste generation patterns globally. Researchers can gain insights into variations, trends, and the reliability of waste estimates.
| Field | Description |
|---|---|
Country |
The name of the country. |
Average Weekly Working Hours |
Average weekly hours worked in the main job. |
Monthly Minimum Wages (USD) |
Minimum monthly wages in USDs. |
Lending Rate (%) |
Lending rates as a percentage. |
Gross Domestic Product (GDP) per Capita (USD) |
GDP per capita in U.S. dollars. |
Human Development Index (HDI) |
A composite index measuring health, education, and standard of living achievements. |
Household Waste Estimate (tonnes/year) |
Estimated household waste generated per year in tonnes. |
Food Service Waste Estimate (tonnes/year) |
Estimated waste generated by the food service sector per year in tonnes. |
Retail Waste Estimate (tonnes/year) |
Estimated waste generated by the retail sector per year in tonnes. |
- Data Cleaning:
- Handled missing values and normalized column names.
- Exploratory Data Analysis (EDA):
- Utilized histograms for understanding data distribution.
- Correlation Analysis:
- Explored correlations between economic indicators and waste generation.
- Categorization:
- Categorized countries based on economic development levels.
- Graphical Analysis:
- Used graphical representations to analyze relationships.
- Regression Analysis:
- Conducted regression analysis for predicting waste generation levels.
- Geospatial Analysis:
- Visualized waste distribution across regions using geospatial maps.
- Sectoral Comparison:
- Compared waste generation levels across sectors.
- Time Series Forecasting:
- Applied time series forecasting for future waste trends.
Based on the provided statistical data, several conclusions can be drawn:
- Average Weekly Working Hours
The average weekly working hours in the considered countries are approximately 38.2 hours, with some variation ranging from 33.2 to 43.3 hours.
- Monthly Minimum Wages
The average minimum wage is approximately 1433.23 USDs. However, there is a significant variation in minimum wages, starting from 170 USDs and reaching 3298.4 USDs.
- Lending Rate
The average lending rate is 6.03%, with a wide range from 1.75% to 20%.
- GDP Per Capita
The average GDP per capita is 57.90 thousand dollars. There is also considerable variation in values, with a minimum of 14.30 thousand dollars and a maximum of 143.30 thousand dollars.
- Human Development Index
The average Human Development Index is 0.894, indicating a high level of development.
- Waste Production
The total waste production in the considered countries varies from 38.8 thousand tonnes to 45.4 million tonnes, with an average of approximately 3.17 million tonnes.
- Waste Estimate by Types (Retail, Household, Food Service)
The analysis provides insights into the distribution of waste production by different types.
In summary, these numerical indicators provide a generalized overview of the economic and social characteristics of the examined countries, as well as their environmental performance in waste production.
In the following correlation matrix, we explore the relationships between key economic indicators and waste generation estimates. The values represent correlation coefficients, ranging from -1 to 1.
| Average Weekly Working Hours | Monthly Minimum Wages (USD) | Lending Rate (%) | GDP Per Capita (USD) | Human Development Index | Household Waste Estimate (tonnes/year) | Food Service Waste Estimate (tonnes/year) | Retail Waste Estimate (tonnes/year) | Total Waste Estimate (tonnes/year) | |
|---|---|---|---|---|---|---|---|---|---|
| Average Weekly Working Hours | 1.00 | -0.63 | 0.22 | -0.48 | -0.61 | -0.22 | -0.18 | -0.18 | -0.20 |
| Monthly Minimum Wages (USD) | -0.63 | 1.00 | -0.48 | 0.74 | 0.88 | -0.02 | -0.05 | -0.05 | -0.04 |
| Lending Rate (%) | 0.22 | -0.48 | 1.00 | -0.44 | -0.56 | 0.02 | -0.02 | 0.01 | -0.00 |
| GDP Per Capita (USD) | -0.48 | 0.74 | -0.44 | 1.00 | 0.73 | 0.07 | 0.13 | 0.09 | 0.10 |
| Human Development Index | -0.61 | 0.88 | -0.56 | 0.73 | 1.00 | 0.10 | 0.09 | 0.06 | 0.09 |
| Household Waste Estimate (tonnes/year) | -0.22 | -0.02 | 0.02 | 0.07 | 0.10 | 1.00 | 0.93 | 0.95 | 0.98 |
| Food Service Waste Estimate (tonnes/year) | -0.18 | -0.05 | -0.02 | 0.13 | 0.09 | 0.93 | 1.00 | 0.97 | 0.98 |
| Retail Waste Estimate (tonnes/year) | -0.18 | -0.05 | 0.01 | 0.09 | 0.06 | 0.95 | 0.97 | 1.00 | 0.98 |
| Total Waste Estimate (tonnes/year) | -0.20 | -0.04 | -0.00 | 0.10 | 0.09 | 0.98 | 0.98 | 0.98 | 1.00 |
In the provided correlation matrix, we explore the relationships between various economic indicators and waste generation estimates. The correlation coefficients range from -1 to 1:
- Strong Positive Correlation (Close to 1): Indicates a direct relationship where an increase in one variable is associated with an increase in the other.
- Strong Negative Correlation (Close to -1): Indicates an inverse relationship where an increase in one variable is associated with a decrease in the other.
- Weak Correlation (Close to 0): Suggests a weak or no linear relationship between the variables.
- Negatively correlated with Monthly Minimum Wages (USD), GDP Per Capita (USD), and Human Development Index.
- Weak positive correlation with Lending Rate (%).
- Strong positive correlation with GDP Per Capita (USD) and Human Development Index.
- Negative correlation with Average Weekly Working Hours.
- Weak positive correlation with Average Weekly Working Hours and weak negative correlation with GDP Per Capita (USD) and Human Development Index.
- Strong positive correlation with Monthly Minimum Wages (USD) and Human Development Index.
- Negative correlation with Average Weekly Working Hours.
- Strong positive correlation with Monthly Minimum Wages (USD) and positive correlations with GDP Per Capita (USD).
- Negative correlations with Average Weekly Working Hours.
- Strong positive correlations among different waste types (household, food service, retail) and total waste estimate.
- Positive correlation with economic indicators, suggesting that more economically developed countries tend to generate more waste.
These findings offer insights into the interplay between economic factors and waste generation patterns, facilitating a comprehensive understanding of the dataset.
In this analysis, we categorize countries into three economic development levels: Low, Medium, and High. The classification is based on specific criteria, and the countries in each category are as follows:
-
Low Economic Development Level:
- Belarus, Bulgaria, Croatia, Greece, Hungary, Latvia, Montenegro, Portugal, Romania, Serbia, Slovakia, Ukraine.
-
Medium Economic Development Level:
- Canada, Czechia, Estonia, Finland, France, Italy, Lithuania, Malta, Poland, Slovenia, Spain, UK.
-
High Economic Development Level:
- Austria, Belgium, Denmark, Germany, Iceland, Ireland, Luxembourg, Netherlands, Norway, Sweden, Switzerland, USA.
The Ordinary Least Squares (OLS) regression results provide insights into the relationship between various economic indicators and the total estimate of waste production in tonnes per year.
- R-squared: 0.156
- Adjusted R-squared: 0.015
- F-statistic: 1.106
- Prob (F-statistic): 0.378
- Log-Likelihood: -618.09
- AIC (Akaike Information Criterion): 1248
- BIC (Bayesian Information Criterion): 1258
-
Constant:
- Coefficient: -5.596e+06
- P-value: 0.930
- Interpretation: The constant term is not statistically significant.
-
Average Weekly Working Hours:
- Coefficient: -1.097e+06
- P-value: 0.163
- Interpretation: The average weekly working hours have a negative impact on the total waste estimate, but the relationship is not statistically significant at the 0.05 significance level.
-
Monthly Minimum Wages (USD):
- Coefficient: -6752.2463
- P-value: 0.055
- Interpretation: Monthly minimum wages show a negative impact on the total waste estimate, and the relationship approaches statistical significance at the 0.05 significance level.
-
Lending Rate (%):
- Coefficient: 1.092e+05
- P-value: 0.842
- Interpretation: Lending rate does not show a statistically significant impact on the total waste estimate.
-
GDP Per Capita (USD):
- Coefficient: 5.984e+04
- P-value: 0.429
- Interpretation: GDP per capita does not show a statistically significant impact on the total waste estimate.
-
Human Development Index:
- Coefficient: 6.289e+07
- P-value: 0.281
- Interpretation: The Human Development Index does not show a statistically significant impact on the total waste estimate.
The overall model does not explain a significant proportion of the variance in the total waste estimate (R-squared = 0.156). Additionally, individual economic indicators do not exhibit strong statistical significance in predicting the total waste production.
It's crucial to note that the large condition number (1.08e+05) may indicate potential issues with multicollinearity or other numerical problems in the model.
To access the answers to the questions, please follow research_question and review the responses and analysis.
To effectively use this dataset, follow the provided analysis steps, adapting them based on your specific research goals and dataset characteristics.
The project is based on the analysis of the relationship between economic indicators and waste generation using a dataset extracted from the 2021 UNEP Food Waste Index Report.
In our exploration of the relationship between economic indicators and global waste generation patterns, we've uncovered several noteworthy insights. Our dataset, encompassing diverse countries, allowed us to observe trends and variations in economic factors and waste production.
-
On average, countries maintain a standard working week of around 38.2 hours, with variations.
-
Monthly minimum wages exhibit substantial diversity, ranging from 170 to 3298.4 USD.
-
Lending rates vary widely, averaging at 6.03%, reflecting the financial diversity among nations.
-
GDP per capita averages at 57.90 thousand USD, showcasing economic differences.
- The HDI indicates a high level of development across the considered countries, with an average of 0.894.
- Total waste production spans from 38.8 thousand tonnes to 45.4 million tonnes, emphasizing the significant environmental impact.
While we express moderate to high confidence in our findings, it's crucial to acknowledge the complexities and potential uncertainties associated with our analysis. The following factors contribute to our level of certainty:
-
Data Limitations: Our findings rely on the available dataset, and any limitations or gaps in the data may influence the accuracy of our conclusions.
-
Assumptions: We made certain assumptions during our analysis, and any deviations from these assumptions could introduce uncertainties in our findings.
-
Sampling Variability: If the analysis involved sampling, there may be inherent variability. Our observations represent patterns in the sample, and variations could exist in a broader population.
This section presents the results of the data analysis conducted using Jupyter Notebook - Analysis
| Technique | Objective | Rationale |
|---|---|---|
Data Cleaning |
Handled missing values and normalized column names for uniformity. | Clean data is essential for accurate analysis, and normalization facilitates consistent comparisons. |
Exploratory Data Analysis (EDA) |
Utilized histograms to understand data distribution. | Histograms provide a visual overview of the data, aiding in identifying patterns and potential outliers. |
Correlation Analysis |
Explored correlations between economic indicators and waste generation. | Understanding correlations helps identify potential relationships and guide further analysis. |
Regression Analysis |
Conducted regression analysis for predicting waste generation levels. | Regression allows for the exploration of relationships between independent and dependent variables, aiding in prediction. |
Geospatial Analysis |
Visualized waste distribution across regions using geospatial maps. | Geospatial analysis provides a spatial context to waste generation patterns, offering a comprehensive view. |
Machine Learning Prediction |
Used linear regression to predict waste generation levels for future years. | Machine learning models offer a quantitative approach to predict trends and patterns in data, contributing to informed decision-making. |
1. Household Food Waste Across Regions: Unraveling the Tapestry of Per Capita Waste and Socioeconomic Disparities
In this section of the project, we aimed to address the first research question: "What is the average per capita household food waste in the studied region, and how does it vary across different income brackets or urban and rural areas?"
- Poland and the USA exhibit similar levels of per capita household food waste, while Germany and Ukraine have moderately elevated values.
- Canada and Denmark show higher estimates, indicating advanced waste management systems, possibly due to effective sorting and recycling programs and responsible population attitudes.
- The United States leads in food service expenditures, with variations observed among countries like Germany, Denmark, Poland, Canada, and Ukraine.
- Diverse distribution hints at evolving trends influenced by consumption strategies and legislative landscapes, emphasizing varying policy needs for food waste management.
- Germany has the lowest retail waste, while Canada and Poland exhibit moderate values.
- The United States and Ukraine show slightly higher retail waste, but Denmark stands out with significantly higher levels influenced by factors like limited local production and higher affluence.
2. Global Economic Impact of Food Waste: Unveiling the Interplay of Economic Consequences and Regional Disparities
In this segment, we sought to delve into the second research question: "What is the collective economic impact of food waste on a global scale, and how does it affect different regions economically?" This exploration involved examining the overall economic ramifications of food waste worldwide and discerning its distinct effects on various regions
- The global average volume is around 3.17 million tonnes, leading to significant economic losses due to devaluation and inefficient resource utilization.
- Higher development level countries like the USA have larger volumes of food waste compared to lower development level countries like Ukraine and Portugal.
- Regression analysis suggests limited impact of indicators like average weekly working hours and minimum wages on predicting food waste volumes, but they are still factors worth considering.
- Countries with higher GDP per capita tend to generate more food waste, emphasizing the link between economic indicators and waste production.
- Global economic impact involves substantial losses, resource inefficiency, and increased expenditures on transportation and storage.
- Disparities in economic development define food waste volumes, and economic indicators interact with induced food waste.
In this segment, our primary objective was to delve into the third research question: "Do correlations exist between the economic prosperity of a region and its success in implementing effective food waste reduction initiatives?" This inquiry was undertaken to uncover the intricate connections between a region's economic well-being and its capacity to execute impactful strategies aimed at reducing food waste.
- Analysis identified variations, emphasizing the need for additional research to fill data gaps, highlighting the importance of comprehensive and reliable data for robust conclusions.
- The model's sensitivity to economic indicators suggests that regions with higher economic development may have more resources and infrastructure for effective waste reduction initiatives.
- Identification of potential risk sources underscores the complexity of the relationship, suggesting optimization of data collection processes and consideration of additional factors.
- The recommendation for supplementary research indicates a commitment to continuous improvement, recognizing the need for exploring evolving dynamics between economic prosperity and food waste reduction.
In conclusion, the analysis provides valuable insights into household food waste, the global economic impact of food waste, and potential correlations between economic prosperity and food waste reduction initiatives, highlighting the importance of further research and a holistic approach for a comprehensive understanding.
- Explanation: Instead of traditional regression, machine learning models like Random Forest or Gradient Boosting could be explored for predictive modeling, potentially capturing non-linear relationships.
- Explanation: To address potential outliers impacting regression results, robust regression techniques like Huber regression could be considered.
- Explanation: Utilizing more advanced geospatial analysis techniques, such as spatial autocorrelation, could provide deeper insights into regional waste distribution patterns.
The findings can be used for further research and development of waste management strategies in the context of economic development.
For a detailed analysis, refer to the Full Project Analysis in the project repository.
analysis_waste - Detailed analysis of waste was conducted, including data type checking, data processing and cleaning, as well as statistical analysis.
Summary analysis-The dataset comprises estimates of household, food service, and retail waste per country, extracted from the 2021 UNEP Food Waste Index Report. It encompasses information on waste generation, confidence levels, and economic indicators such as GDP per capita, lending rates, and more.
Predictions- The analysis focuses on addressing data gaps, sensitivity scenarios, and risk sources, offering solutions for improvement
While the chosen analysis techniques have provided valuable insights, it's crucial to acknowledge potential flaws and consider alternative approaches to enhance the robustness of our findings. Adapting methods based on ongoing discoveries and advancements in data science ensures a dynamic and thorough analysis approach.
Answers for Actionable Research Questions.
Please be aware that the analysis is currently underway, and any updates or improvements will be incorporated into future milestones.
What is the average per capita household food waste in the studied region, and how does it vary across different income brackets or urban and rural areas? What is the collective economic impact of food waste on a global scale, and how does it affect different regions economically? Are there correlations between the economic prosperity of a region and its success in implementing effective food waste reduction initiatives?