Python notebooks for Harvard APMA 207

Archived repo of data science Jupyter Notebooks (machine learning, statistical modeling, data analysis)

NYU Core UA 107 lecture demonstration notes

Jupyter notebook presenting the process of data preparation, research question formulation, data analysis, and data modelling with the goal of extracting insights from the 2018 PISA Dataset

Repository with code and notebooks used to train myself in statistics, probability and machine learning using publicly available datasets.

The repository stores data-oriented Jupyter Notebooks. The provided examples include well-studied ML models and are used for financial modelling. The presented models include Logistic Regression, Kernel SVM, PCA, Random Forests, Body/tail calibration using MLE etc.

Analysis of bus delays data from TTC (Toronto Transit Commission). So far we concluded that there is a statistically signigificant difference in number of delays between days with good weather and bad weather within the "Mechanical" and "General" delay reasons. More details in Jupyter Notebooks.

In collaboration with Google, Udacity provides an introductory course to A/B testing. The course covers the experiment design and test results analysis approaches. This report and notebook provide a walkthrough of the course's final project, including choosing and characterizing metrics, determining experiment size and duration, and analyzing te…