13-part master curriculum: NumPy vector math, Pandas wrangling, high-performance grouping, data cleaning, reshaping, descriptive statistics, EDA techniques, visualization, correlation, statistical hypothesis testing, memory optimization, and end-to-end capstone workflow.
Foundations of data analytics: four types of analytics (descriptive, diagnostic, predictive, prescriptive), analytics vs data science, business question formulation, and the analytics lifecycle.
Core scientific computing with NumPy: N-dimensional array memory layouts (strides, C vs Fortran order), vectorized arithmetic, broadcasting rules, advanced indexing, and linear algebra routines.
Essential tabular data manipulation in Pandas: Series and DataFrame structures, index mechanics, label-based (loc) vs position-based (iloc) selection, data type inspection, and filtering.
Efficient data transformations: vectorization vs apply/itertuples bottlenecks, groupby split-apply-combine mechanics, custom aggregation functions, window transforms, and filtrations.
Production data cleaning workflows: identifying and handling MCAR/MAR/MNAR missing values, deduplication, string parsing, type casting, and IQR/z-score outlier detection and treatment.
Mastering tabular geometry: wide vs long formats, pivot tables, melt un-pivoting, hierarchical multi-index stacking/unstacking, cross-tabulations, and Tidy Data design principles.
Core statistical summaries: mean vs median vs trimmed mean, variance and standard deviation, skewness, kurtosis, empirical rule (68-95-99.7), and fundamental probability distributions.
Structured EDA methodology: univariate distribution analysis, bivariate relationships, multivariate interaction discovery, segment comparisons, and data-driven hypothesis generation.
Visualizing analytical patterns: Matplotlib figure/axes architectures, Seaborn statistical plots, distribution histograms/KDEs, box/violin comparisons, scatter matrices, and heatmaps.
Quantifying variable associations: sample covariance math, Pearson linear correlation, Spearman rank correlation, Kendall tau, correlation heatmaps, and confounding variable pitfalls.
Drawing statistically rigorous conclusions: sampling distributions, Central Limit Theorem, confidence intervals, null hypothesis testing, p-values, t-tests, ANOVA, and Chi-Square tests.
Scaling analytics on single-machine RAM: memory profiling, numeric dtype downcasting, Categorical string compression, chunked streaming, Parquet vs CSV formats, and PyArrow integration.
Complete professional analytics capstone: translating ambiguous business questions, data hygiene audits, rigorous multi-dimensional EDA, hypothesis testing, and executive insight delivery.