data-science-expert

Exploratory data analysis in pandas involves handling missing values, removing duplicates, and filtering outliers to prepare trustworthy datasets. This Skill guides end-to-end data cleaning and EDA to reduce noise and ready tabular data for modeling.

Feature engineering for machine learning in Python includes building interaction and polynomial features, binning numeric variables, encoding categoricals, and scaling predictors. This process transforms raw tabular data into modeling-ready datasets with informative features.

Time series stationarity checks and ARIMA forecasting in Python require decomposing time series data and fitting models to predict future values. This Skill handles stationarity tests, decomposition, and ARIMA fitting for raw time-series datasets.

You can run A/B tests using t-tests and proportion z-tests to evaluate experiments within your dataset. This Skill provides statistical workflows to validate experiment columns and generate statistical conclusions from modeling-ready data.

You need Python-based pandas and numpy to transform data for visualizing correlations and distributions. This Skill uses common plotting libraries alongside these dependencies to provide visualization guidance for exploratory analysis.

Statistical inference components for evaluating experiment data include t-tests, proportion z-tests, and correlation analysis. This Skill applies these statistical tests to validate experiments and derive reliable insights from cleaned datasets.