The Differential Impacts of Human Capital and Infrastructure on Sustainable Development: An Empirical Analysis
This study looks at country-level data to explore the dynamics among human capital, infrastructure, and a country’s progress toward the United Nations Sustainable Development Goals (SDGs). Utilizing the confirmatory factor analysis method, I develop a new Infrastructure Index and combine it with the World Bank’s dataset on Human Capital Index to evaluate the relative impact of these factors on a country's SDG scores. My findings affirm the integral roles of both human capital and infrastructure in the sustainable development context. However, a stronger correlation between human capital and the SDG Index suggests that policymakers seeking to advance the sustainability agenda should prioritize investments in human capital over infrastructure. Moreover, the study uncovers nuanced relationships between these indicators and specific SDGs. Human capital has a significant association with SDG 5 (Gender Equality), whereas infrastructure does not. Both human capital and infrastructure affect SDG 1 (No Poverty), with no statistical difference between their effects. Interestingly, while human capital correlates more strongly with SDG 13 (Climate Action), this relationship is negative due to the larger carbon footprint of more developed economies. These findings can inform policy decisions for goal-specific sustainable development strategies.
Capstone Project for MIT Applied Data Science Program
The pricing and supply of new cars are generally predictable, controlled by Original Equipment Manufacturers (OEMs), with the only variations occurring due to dealership-level discounts at the final stage of the customer journey. However, the used car market operates differently. It is characterized by significant uncertainties in both pricing and supply. The valuation of a used car can be influenced by multiple factors including mileage, model year, brand, and overall condition. For sellers, setting an accurate price is challenging yet crucial for market competitiveness. In this data sience project I develop a robust pricing model that can predict the prices of used cars with 94% accuracy. Such a model can both facilitate fair market transactions and enable the company to implement differential pricing strategies that could drive profitability and market growth.
Capstone Project for IBM Professional Certificate in Data Science
In this project, I constructed a machine learning model that can predict whether Falcon 9 will land successfully in the first stage. Falcon 9 is classified as a medium-lift partially reusable rocket, used to launch hefty communications and satellites into Earth orbit or ferry austronaouts to and from the International Space Station. As of April 2022, SpaceX offers Falcon 9 rocket launches for USD 62 million, which means around USD 1,200 per pound of payload. For comparison, per pound cost of SpaceX competitors is 3 to 5 times more expensive, whereas traditional NASA space shuttles, retired in 2011, cost an average of USD 1.6 billion per flight. SpaceX is able to provide rocket launches for unprecedented low prices, because it can reuse the first stage, which significantly reduces the demand for new cores. Therefore, determining whether the first stage will land, helps to estimate the cost of a launch. This information can also be used if another company wants to bid against SpaceX for a rocket launch. For this project, I use API requests to pull data from https://api.spacexdata.com/v4 and webscraping to collect data from Wikipedia.