Saqib Fayaz

Logo

View the Project on GitHub saqibfayaz/saqibfayaz1.github.io

Data Analyst

Porfolio Projects Directory

Technical Skills: Python, R, NumPy, Pandas, Matplotlib, Seaborn, SQL, Tableau, Power BI, Google BigQuery, SKLearn, MS Excel, Git, HTML, CSS, JavaScript

Other Skills : Economics, Psychology, Data Preprocessing, Data Wrangling, Data Plotting, Linear Regression Modeling, Detail-Oriented

Education

Work Experience

Web Developer @ MyTravaly (September 2019 - June 2021)

Projects

Telco Customer Churn Analysis and Retention Strategies

Github

In this project, I analyzed customer churn data to identify factors influencing customer retention. I used Python along with libraries like Pandas and NumPy for data cleaning and manipulation, Matplotlib and Seaborn for visualization, and SciPy for statistical testing. The analysis involved exploring relationships between customer attributes such as gender, online security, and churn status. Although a chi-square test showed no significant association between gender and churn, further analysis revealed that customers without online security were more likely to churn, with 20% of women and 21% of men unsubscribing. This project demonstrates how exploratory data analysis and statistical methods can provide actionable insights for addressing business challenges like customer churn.

Analysis of Y Combinator-Funded Startups

Github

Analyzed and visualized factors associated with the success and failure of 4,845 Y Combinator-funded startups using Python libraries (Pandas, Matplotlib, Plotly, Seaborn). Merged datasets on companies, industries, regions, and founder education to create a comprehensive dataset, transforming categorical data into numerical representations with cat.codes to identify correlations. Explored success factors such as industry, region, and team size, defining success as startups with 100+ employees and active status. Visualized findings with stacked bar charts, tree maps, and sunburst plots to compare trends between successful and failed startups, uncovering regional and sector-specific insights into startup outcomes.

1

Analysis of Bitly User Data for USA.gov

Github

Analyzed and visualized user data on government-related URLs shortened via Bitly, using Python libraries (Pandas, Seaborn, Matplotlib) to extract and process JSON data. Identified top time zones and usage patterns, categorized users by OS type (Windows vs. non-Windows), and grouped data by time zone and OS for detailed insights. Normalized and visualized data to highlight regional and platform-specific trends in service activity.

3

Fandango Score Comparison

Github

Analyzed movie ratings across multiple platforms to identify discrepancies and correlations, focusing on Fandango and Metacritic ratings. Applied statistical methods to calculate central tendencies and performed regression analysis using SciPy’s stats module, evaluating model accuracy with R-squared, mean absolute error, and root mean squared error. Visualized linear regression results and detected outliers, showcasing expertise in data cleaning, statistical analysis, and regression modeling.

4

Certifications