These are some of the most exciting projects I have worked on during my time at UCSD.

The effects of increased trade and broadband access on the development of Iran, November 2015. This paper uses the Development Accounting methodology, developed by Francesco Caselli in his paper “Accounting for Cross-Country Income Differences” (2005) to assess the impact of the increased trade and broadband access in Iran due to a possible lifting of sanctions.

Remote sensing of urban areas destroyed by war using Landsat data, March 2016. In this paper I use ArcGIS and the relatively new tool Google Earth Engine to understand whether it is possible to detect destruction in urban areas by measuring the reflectance values of those areas. The hypothesis was that this destruction increases reflectance by scattering rubble all over an area. Both cross-section and time-series reflectance data collected using Google Earth Engine in the city of Homs supported this hypothesis.

Estimating the impact of bilateral and multilateral trade agreements, March 2016. In this paper I use the gravity model, the workhorse of trade economics, together with fixed-effects and random-effects estimation to assess the differential impact of multilateral vis-à-vis bilateral trade agreements.

Leveraging mobile money for education in Benin: pre analysis plan, March 2016. This paper was a class exercise in drafting a pre-analysis plan of a randomized controlled trial. The real RCT project, developed by Dr. Jen Burney, was to install mobile-money platforms in schools in Benin so they could use this platform to receive tuition-specific remittances. This pre-analysis plan goes through the issue of causal chain, proper identification of the effect, randomization, and power calculations. Dr. Burney authorized the use of her project as the subject of this exercise but does not endorse its contents.

Predicting hospitalization using patient-level data, June 2016. In this paper I compare the performance of a linear model, the multivariate logistic model, versus the Random Forest estimator in predicting hospitalization using patient-level health data.

Remote sensing vegetation and biomass using LiDAR systems, June 2016. ICESAT-2, a laser altimetry satellite, was developed to detect minute changes in ice sheets and ice-related structures around the globe. It also has another potential use, because the analysis of the returning waveform can yield much information about areas covered with vegetation. This paper is a review of the methods used to retrieve vegetation information from these waveforms, and I conduct original analysis of using ICESAT-2 waveforms in MATLAB to demonstrate these methods.