Nigar (Leeza) Sultana
Business Analyst with 3+ years of experience in using statistical and ML modeling to make data-driven strategies for businesses. Proficiently skilled in employing Python, R, STATA, SQL Query, and Tableau to extract actionable insights to facilitate informed business decisions.
View My LinkedIn Profile
Selected projects of Statistical Modeling, Data Analysis, and Visualization
Modeling Loan Default in Peer-to-Peer Lending Services
A discrete-time survival- analysis was built to predict the loan default using Lending Club’s loan data, including customer information on loan purpose, grade, annual income, FICO score, verification status, etc.
View the full description and code
The Effect of Country of Manufacture on Fly Reels’ Prices
Used Hedonic price theory and Oaxaca Decomposition to find consumers’ willingness to pay for an American-made-fly-fishing reel than a foreign one
View the full description and code
Predicting Customer Response in Bank Marketing Campaign for Term Deposit Subscription
Four classification models, such as decision tree, logistic regression, random forest, and support vector machine, were used to predict the customer response in a direct bank marketing campaign. Th eprimary focus was on the predictive capabilities of these models, followed by a comparative analysis to identify the best-performing model.
View the full description and code
Analyzing Annual Earnings of H1B Applicant Business Analysts
The H1B visa is a temporary nonimmigrant visa permitting American employers to engage highly educated foreign professionals, primarily within the domains of Science, Technology, Engineering, and Mathematics(STEM). This project explores the typical earnings of an H1B applicant for a business analyst position in different regions in the USA— and the value of those earnings after adjusting for the cost of living across the USA.
View the full description and code
Exploratory Data Analysis of a Dog Assessment Company
Dognition is a company that offers a unique understanding of dogs’ personalities and capabilities through a comprehensive “Dognition Assessment” provided by three subscription plans, consisting of 20 games designed by experts to categorize dogs into nine profile types. Once the assessment is completed, dog owners receive a personalized report of 10-15 pages detailing their dog’s cognitive strategies and game-specific results. The company’s primary objective is to collect comprehensive data from diverse dog breeds. To achieve this, the company seeks to identify potential business changes to boost the number of completed tests on its website. This analysis recommends business changes that Dognition can implement to increase the number of tests that users complete on its website.
View the full description and code
Income, Well-being, and Happiness: A State-Level Analysis
Living in the United States offers diverse experiences, with each state presenting unique financial, educational, healthcare, cost of living, infrastructure, and mental well-being perspectives. While the choice of where to live is often subjective, there are objective measures that can aid in state comparisons. Median income, coupled with living expenses, provides valuable insights into the living standards of residents. Furthermore, a state’s commitment to ensuring its citizens’ well-being is pivotal. Additionally, while happiness is a subject of psychological discourse, achieving a certain income level enables individuals to access essentials like healthcare and education, contributing to emotional well-being. Crucial elements such as poverty rates, cost of living, public safety, and environmental quality significantly impact people’s quality of life.
This data story analyzes state-level data from various sources to evaluate income, well-being, and happiness across all states.
Infographic for UNDP Accelerator Lab: A Volunteer Project for Viz for Social Goods
In 2019, UNDP’s Accelerator Labs were built to learn what works and what doesn’t in sustainable development to achieve the Sustainable Development Goals (SDGs) in time. One of the three team members of the Accelerator Labs, the solution mapper, has discovered 359 grassroots energy solutions from different regions, demographics, and energy sources in the world. These solutions are promising for UNDP’s Strategic Plan (2022 – 2025), which sets out the ambitious objective of increasing access to clean and affordable energy for 500 million people by speeding up investment in distributed renewable energy solutions. This volunteer project (a Viz for Social Good project) aims to build an interactive dashboard and infographic for the government partners, the international donor community and the general public to advocate for local solutions in achieving SDGs.
View the full description and code