Riya Ambekar

Detail-oriented and collaborative aspiring data analyst with a strong foundation in data analysis, statistical modelling, and machine learning using R and Python. Skilled in hypothesis testing, regression analysis, and predictive modelling to extract actionable insights and support evidence-based decision-making. Proficient in data wrangling, visualization, and clearly communicating complex findings to both technical and non-technical audiences. Confident in using Power BI to design and manage interactive dashboards and reports, with additional experience in Microsoft Excel for reporting and analysis. Eager to learn and expand skills in SQL to enhance data querying and analysis capabilities. Demonstrated ability to work independently, manage priorities effectively, and contribute to team-based analytical projects. Committed to continuous learning and growth in applied statistics, statistical computing, and modern data tools.

A Statistical Analysis of Old Age Homes (2022)

Objective: To analyze the quality of life among senior citizens and examine both the positive and negative aspects of old age homes.
Key Contributions:

• Utilized three primary datasets with sample sizes of 114, 148, and 110 participants.
• Applied a range of statistical methods, including Linear Regression, Ordinal Logistic Regression, and both parametric and non-parametric hypothesis testing.
• Implemented machine learning techniques and statistical tests to extract actionable insights and support evidence-based conclusions.
• Interpreted results to highlight patterns and factors influencing well-being in elderly populations.

Robust Discriminant Analysis (2025)

Objective: To investigate the impact of robust statistical methodologies on the performance of Quadratic Discriminant Analysis (QDA) and Soft Independent Modeling of Class Analogy (SIMCA).
Key Contributions:

• Explored the application of robust statistics to enhance the reliability of classical discriminant analysis methods in the presence of outliers and non-normality.
• Utilized a motivating dataset aimed at distinguishing between naturally occurring earthquakes and man-made explosions.
• Assessed model performance under classical and robust frameworks, comparing classification accuracy and sensitivity to noise.
• Demonstrated the practical advantages of robust approaches in geophysical data classification tasks.