CEU Electronic Theses and Dissertations, 2019
| Author | Al-Shaibani, Ahmed Khaled Saif |
|---|---|
| Title | A Prediction Model of Foreign Aid Projects Funded by the United States: An Analysis Using Data from Mexico |
| Summary | The purpose of this research is to provide policy-makers with a tool to alleviate some of the burden resulting from policy making in Foreign Aid. Using data from the foreign aid projects funded by the United States in Mexico in 2017, the purpose is to predict the probability of project completion given certain characteristics. The analysis is carried out using a Logistic Regression with 5-fold cross validation. The model has an AUC of 93.10%, which means that when the model is shown two randomly selected projects, one completed and one not, 93.10% of the time it will assign a higher probability of project completion to the project that is actually completed. The findings show that a typical project that is likely to be completed is done in the area of Peace and Security with funding arriving in the second quarter of the fiscal year. Due to security reasons, the organization type is retracted in the data. The share of completed projects with such characteristics is 94%. In contrast, a project that is typically less likely to be completed is done in the Environmental sector with funding arriving in the third quarter of the fiscal year. The organization type is in the NGO sector. The share of completed projects with such characteristics is 18%. Findings generated from this analysis can potentially be used by policy-makers in prioritizing which policies to allot for projects financed by foreign aid. |
| Supervisor | Békés, Gábor |
| Department | Economics MA |
| Full text | https://www.etd.ceu.edu/2019/alshaibani_ahmed.pdf |
Visit the CEU Library.
© 2007-2021, Central European University