This project implements in Python some common statistical analysis methods used in data analysis, including Entropy, Mutual Information, Kolmogorov–Smirnov test, Kullback-Leibler divergence (KLD), AB tests (Mann-Whitney U and t-tests)

Tabular Data Drift Detector & Reporter:
A CLI tool that connects to any database or CSV, computes statistical drift (KS-test, Jensen-Shannon) between “baseline” vs. “current” data, and emits a Markdown/HTML report with charts.

The PKS Test software was designed to directly verify whether a phylogeny is clock-like following a Poisson law.

A natural time analysis of the Earthquake Cycles in Taiwan by evaluating EPS scores using R and Python.

A basic report and code to apply Kolmogorov Smirnov tests on a given set of data using Jupyter Notebook.

It's an open-source engine that scores and stress-tests machine learning models for reliability, robustness, and hidden failures before deployment.

A Hemispherical Comparison code using HEALPiX and Kolmogorov–Smirnov test for probing isotropy

OOP project in C++ implementing various random number generators (ICG with KS test, MRG, LFG, Mid-Square) with user-input parameters for sequence generation and randomness testing.

A system for monitoring statistical data distribution shifts in distributed settings

Applied KS test and T-test to check whether rental subsidy rate’s distributions are different across different PHAs and implemented Pearson-correlation analysis to explore the linear correlation between rental subsidy rate and other factors.

Analysis of far-infrared and X-ray properties of quasars using astronomical catalogues and statistical tests.

Drift monitoring API for ML models — FastAPI backend, Streamlit dashboard. Detects feature drift via KS-test, PSI, and IQR fences, with automatic column profiling and human-in-the-loop schema verification. Validated on 4.5M rows: Precision 1.000, Recall 0.939, F1 0.969 (production); 100% detection on synthetic drift.

Detect data and prediction drift in production ML models. Pure-NumPy detectors (PSI, KS, JS, prediction shift) with webhook/email alerts and a Rich dashboard.

Probability and Statistics for Data Scientist - Course Project. Various Testing(Hypothesis, T-Test, Z-Test, KS Test etc) are performed on Covid and crime dataset for Los Angeles. Auto Regression and Linear Regression are also performed without using scimitar libraries in python.