4.1 principal component factor
As illustrated in Figure 4, positive factor loading coefficients for principal component one are population growth (annual percent), high-tech exports (current dollars), high-tech exports (percent of manufactured exports), total gross public expenditures on education (percent of GDP), net imports of energy (percent of energy use), energy use (kilograms of oil equivalent per capita), and all tariff lines in the tariff schedule that reach the highest international tariff rates. Share of products (%). These indicators cover population growth, economic structure, investment in education, energy use and trade policy. Considering these indicators together, a new indicator can be proposed: a composite index of sustainable development. This composite index would cover a combination of population growth rate, level of high-tech industries, educational inputs, energy utilization efficiency and trade policies, and would aim to assess a country's overall performance on a sustainable development path.
The principal component two-factor factors with positive factor loading coefficients are high-tech exports (current dollars), change in inventories (constant local currency units), Gini (GINI) coefficient, gross domestic savings (current local currency), high-tech exports (as a percentage of manufactured exports), ore and metal imports (as a percentage of merchandise imports), energy use (kilograms of petroleum equivalent per capita), claims on the rest of the domestic economy ( annual growth rate as a percentage of broad money), other tax revenues (units of local currency at current prices), and tax revenues (as a percentage of gross national product (GDP)). These indicators cover a wide range of data on the country's economy, trade, resource utilization, revenues and expenditures. Considering these indicators together, it is possible to propose a new indicator: a comprehensive index of economic sustainability. This index would take into account data on a number of aspects, such as the level of a country's high-tech industry, resource utilization, financial situation, trade structure and balance of income and expenditure, in order to assess the sustainability of a country's economy.
The principal components three-factor factors with positive factor loading coefficients are population growth (annual percent), change in inventories (constant price local currency units), Gini (GINI) coefficients, bound tax rates, simple mean, all products (percent), nurses and midwives (per 1,000 population), net energy imports (as a percent of energy use), energy use (per capita kilograms of oil equivalent), other taxes (current price local currency units), and the share of all products in the tariff schedule that reach the international maximum tariff rate (percent). ), and the proportion (%) of products in the tariff schedule that reach the highest international tariff rate for all tariff lines. These indicators cover data on population growth, economic structure, income inequality, tax policy, energy use, and many other aspects. Considering these indicators together, it is possible to propose a new indicator: a composite socio-economic development index. This index would take into account data from a number of aspects, such as population growth rate, stability of economic structure, fairness of income distribution, rationality of tax policy, efficiency of energy utilization, etc., in order to assess the overall situation of socio-economic development in a country.
Positive factor loading coefficients for the four factors of the principal components are high-tech exports (current dollars), computer, communications, and other services (percent of imports of business services), bound tax rates, simple mean, all products (percent), gross domestic savings (current local currency), high-tech exports (percent of manufactured exports), claims on the rest of the domestic economy (annual rate of growth in percent of broad money ), other tax revenues (in units of current local currency), tax revenues (as a percentage of gross national product (GDP)), and the share of products in the tariff schedule that reach the highest international tariff rates for all tariff lines (in percent). These indicators cover a wide range of data on the structure of the country's economy, trade, savings and debt, and tax policy. Considering these indicators together, it is possible to propose a new indicator: a composite index of economic openness. This index would take into account data from a number of aspects, such as the proportion of high-tech exports, the proportion of trade in services, the level of tariffs, the savings rate, the tax situation and the growth of debt, in order to assess the degree of openness of a country's economy and its dependence on foreign trade.
Figure 5(a) shows the relationship between the number of factors and the characteristic root. By dimensionality reduction of the metrics, we simplified the 17 metrics into 4 main components. The main reason for choosing these 4 metrics is that the figure clearly shows that when the number of factors is 4, the graph line tends to stabilize and the trend of decreasing eigenroots becomes obvious. As shown in Figure 5(b), the differences in the explanation of the total variance of the four components we selected are significant. The cumulative variance explained by these four indicators reaches 77.828%, indicating that they can better reflect most of the information of the original 17 indicators, and thus the other indicators can be discarded to simplify the model and improve processing efficiency. Although choosing five indicators may cover more data, it will significantly increase the complexity of modeling operations. In order to maintain the simplicity of the model and ease of operation, we finally decided to use four indicators.
Figure 6(a) shows the heat map of the share of each indicator in the principal component analysis. In this heat map, the darker color indicates the higher percentage of the indicator in the principal components. From the figure, we can see that indicator A1 has a larger share in T1, T2, T8, T11, T13, and T17; while A2 has a significant share in T16. Since most of our data obeyed a normal distribution, and although a few data did not exactly match, they showed an overall trend of normal distribution, we used Pearson correlation analysis. components, and the lighter color indicates the stronger negative correlation. As can be seen from the figure, there is a strong negative correlation between A2 and Y4, while the positive correlation between A2 and Y2 is more significant compared to the other combinations.
Figure 6(b) shows a heat map of the correlation between the indicators and the principal components. In this graph, the darker color indicates the stronger positive correlation between the indicators and the principal
Figure 7 is a four-quadrant scatterplot reflecting the distribution of the indicators in each principal component, showing the relationship between the factors' roles in the principal components and indirectly revealing the relative importance of each indicator. Table 2, on the other hand, lists in detail the items corresponding to each factor of T1-T17, Y1-Y4, and each principal component of A1-A4, respectively, providing a clear reference for further analysis.
In the Analytic Hierarchy Process (AHP), we use a scale of 1 to 5 to indicate the relative importance of two factors: where 1 means that both elements are equally important; 3 means that one element is slightly more important; and 5 means that one element is significantly more important. 2 and 4 are used to indicate the median of the neighboring judgments. Based on this scale, we conducted a two-by-two comparison and used a three-person evaluation team to take the mean value to determine the comparison matrix.
Considering that the development of a two-by-two comparison matrix involves professional judgment and data analysis, especially when dealing with complex illegal wildlife trade issues, the lack of specific research data and expert opinions can make it challenging to provide precise values. Therefore, we assumed a relatively reasonable two-by-two comparison matrix (Figure 8) based on general knowledge of the illegal wildlife trade problem for further analysis. We considered several key factors: first, sustainability is critical to reducing dependence on wildlife, and illegal trade can be radically reduced by providing sustainable livelihood alternatives. Second, economic sustainability can help communities and countries resist the temptation to trade in illegal wildlife by providing long-term economic benefits and reducing the pursuit of short-term gains. In addition, the capacity for socio-economic development indirectly influences attitudes and demand for illegal trade by raising education levels, increasing civic awareness and providing alternatives to economic activity. Finally, while economic openness may increase opportunities for cross-border trade in illegal wildlife products, it also opens up possibilities for international cooperation and the implementation of more effective regulatory measures. Together, these factors have shaped our consideration of strategies aimed at addressing illegal wildlife trade in a multifaceted and integrated manner.
Table 3:Items represented by each principal component in Figure 7
Number
|
Program
|
T1
|
Population growth (annual percentage)
|
T2
|
High-tech exports (current dollars)
|
T3
|
Change in inventories (constant local currency units)
|
T4
|
Gini coefficients
|
T5
|
Computer, communication and other services (percentage of imports of business services)
|
T6
|
Bound tax rates, simple mean, all products (percentage)
|
T7
|
Gross domestic savings (current local currency)
|
T8
|
High-tech exports (percentage of manufactured exports)
|
T9
|
Total public expenditure on education (% of GDP)
|
T10
|
Nurses and midwives (per 1,000 population)
|
T11
|
Net energy imports (percentage of energy use)
|
T12
|
Ore and metal imports (percentage of commodity imports)
|
T13
|
Energy use (kilograms of oil equivalent per capita)
|
T14
|
Claims on the rest of the domestic economy (annual growth rate as a percentage of broad money)
|
T15
|
Other taxes (local currency units at current prices)
|
T16
|
Taxes (% of gross domestic product (GDP))
|
T17
|
Proportion of products in the tariff schedule that reach the highest international tariff rate for all tariff lines (%)
|
Y1
|
Estimated number of elephants illegally killed per year
|
Y2
|
Wildlife trade shipments rejected or found to be illegal as a percentage of total shipments
|
Y3
|
Number of world wildlife seizures
|
Y4
|
Number of poaching incidents in Africa
|
A1
|
Sustainability composite index
|
A2
|
Composite economic sustainability index
|
A3
|
Composite index of socio-economic development
|
A4
|
Composite index of economic openness
|
Table 3 showed items represented by each principal component in Figure 7.The calculation of the weights of the Analytic Hierarchy Process (AHP) showed that the weight Based on the results of these analyses, we can see that in the strategy to combat illegal wildlife trade, the enhancement of sustainable development capacity is particularly important, followed by the enhancement of economic sustainability and socio-economic development capacity. Economic openness, although important, is relatively less important among these factors. This weighting distribution guides us to focus on enhancing sustainable development capacity when formulating strategies to achieve long-term and fundamental impacts. In developing national policy programs to reduce illegal wildlife trade based on the evaluation of the indicator system, governments should focus on enhancing sustainable development capacity, economic sustainability, socio-economic development capacity, and increasing economic openness. Based on these priorities, we propose specific national policy programs and the evaluation of their expected effects (Figure 9).The implementation of these policy programs will require coordinated cooperation between various sectors of the national government, as well as close cooperation with non-governmental organizations and the private sector. Through these integrated measures, illegal wildlife trade can be radically reduced, thereby conserving biodiversity and contributing to sustainable economic and social development.
Figure 10 shows the prediction of the quantity of illegal wildlife trade over time based on real data versus the implementation of the policy program. As can be seen from the figure, the real data show that illegal wildlife trade is on an upward trend, and the volume of illegal trade increases year by year. In the prediction of ARIMA time series model, the MAPE value is 0.067398%, which is much lower than the linear regression model, showing that the percentage error between the predicted value and the actual value of ARIMA model is extremely small, and its RMSE value is 0.072324, which is significantly lower than that of the linear regression model, indicating that the ARIMA model has less error in prediction. Although the R-squared value of ARIMA is 0.66939, which is lower than the linear regression model, showing that its fitting effect is not as good as the linear regression model. Whereas the linear regression model has a relatively high MAPE value of 1.5278% and a root mean square error (RMSE) value of 0.62662, its R-squared value of 0.84305 indicates that the model explains 84.305% of the variance of the target variable better, which usually indicates that the model is relatively well fitted.Thus, curbing illegal wildlife trade requires Fig.10 Prediction of the Illegal Animal Trade Assessment Index by the ARIMA model with time growth a long-term commitment and sustained effort.
Although the results may not be satisfactory in the short term, adopting the right strategies and sustained policy implementation are essential for the conservation of wildlife and their natural habitats. To ensure that these efforts produce tangible results, a comprehensive strategy of international cooperation, scientific approaches and community involvement is needed. In the long run, conserving wildlife is not only responsible for biodiversity, but also for the future of human existence. Therefore, despite the many challenges, we must persevere and continue to optimize and adjust our strategies with a view to achieving the ultimate conservation goals.