Projects

FICO scores are typically the "Go-To" method in taking preventative measures for loss provision and are utilized to provide a good indication of how likely a customer is to default on their mortgage.

Quantitative Puzzle involving Markov-Chains

Risk managers will typically collect data on the loan borrowers. The data will usually be in a tabular format, with each row providing details of the borrower, including their income, total loans outstanding, and a few other metrics.

A data preprocessing pipeline streamlines the complex process by automating a series of steps, further enabling data professionals to efficiently and consistently preprocess diverse datasets.

​

​

Many individuals limit their use of social media, in which I myself oftentimes completely avoided it in the past during times of important exams. Regardless, content creators must still be able to determine the optimal time to take advantage of making content and when not too. Thankfully, Instagram Reach Forecasting can help content creators do this effectively with a little added help from Python!

When trading natural gas storage contracts, it is important that the available market data must be of higher quality to enable the instrument to be priced accurately. This entails extrapolating the data available from external feeds to provide more granularity, considering seasonal trends in the price as it relates to months in the year. To price the contract, we will need historical data and an estimate of the future gas price at any date.

We will be creating a script that can be used to price a Commodity Storage Contract. To do this we will incorporate engineering, risk, and model validation to incorporate this model into production code.

Utilized Quantitative Balance Sheet Strategy of e-merchant data to:

-Create a valuation document for a fintech loan portfolio.
-Analyze historical data and compute repayment rates.
-Forecast cash flows in R to find the portfolio's present value.