WGU D600 Task 1: Multiple Linear Regression
Analysis
Course: D600 – Statistical Data Mining
Assessment: Task 1 – Multiple Linear Regression
Analysis (Performance Assessment)
Academic Year: 2026/2027 Updated
Section 1: Executive Summary & Problem
Statement
1.1 Business Context and Objective
In modern data-driven enterprises, understanding
the quantitative relationships between operational
drivers and a continuous target metric is crucial for
strategic decision-making. The primary objective of
this project is to construct, evaluate, diagnose, and
refine a Multiple Linear Regression (MLR) model
using Python. This model predicts a continuous
response variable (𝑌) based on multiple explanatory
predictor variables (𝑋1 , 𝑋2 , … , 𝑋𝑝 ).
1.2 Scope of the Report
This report documents the complete data mining
workflow, adhering to rigorous statistical standards:
, 1. Data Preparation & Exploratory Data
Analysis (EDA): Handling missing values,
encoding categorical features, and exploring
bivariate relationships.
2. Model Specification & Ordinary Least
Squares (OLS) Estimation: Fitting the baseline
model using Python (statsmodels and scikit-
learn).
3. Assumption Verification: Rigorously testing
OLS assumptions (linearity, homoscedasticity,
normality of residuals, and
independence/multicollinearity via Variance
Inflation Factors).
4. Model Diagnostics & Reduction: Identifying
high-leverage points, outliers, and influential
observations (Cook’s Distance), followed by
feature selection to optimize performance.
5. Business Interpretation &
Recommendations: Translating statistical
coefficients into actionable business intelligence.
Section 2: Data Preparation & Exploratory Data
Analysis (EDA)
2.1 Python Environment & Setup
To execute the data mining workflow, we utilize
standard Python scientific libraries: pandas, numpy,
matplotlib, seaborn, and statsmodels.
Analysis
Course: D600 – Statistical Data Mining
Assessment: Task 1 – Multiple Linear Regression
Analysis (Performance Assessment)
Academic Year: 2026/2027 Updated
Section 1: Executive Summary & Problem
Statement
1.1 Business Context and Objective
In modern data-driven enterprises, understanding
the quantitative relationships between operational
drivers and a continuous target metric is crucial for
strategic decision-making. The primary objective of
this project is to construct, evaluate, diagnose, and
refine a Multiple Linear Regression (MLR) model
using Python. This model predicts a continuous
response variable (𝑌) based on multiple explanatory
predictor variables (𝑋1 , 𝑋2 , … , 𝑋𝑝 ).
1.2 Scope of the Report
This report documents the complete data mining
workflow, adhering to rigorous statistical standards:
, 1. Data Preparation & Exploratory Data
Analysis (EDA): Handling missing values,
encoding categorical features, and exploring
bivariate relationships.
2. Model Specification & Ordinary Least
Squares (OLS) Estimation: Fitting the baseline
model using Python (statsmodels and scikit-
learn).
3. Assumption Verification: Rigorously testing
OLS assumptions (linearity, homoscedasticity,
normality of residuals, and
independence/multicollinearity via Variance
Inflation Factors).
4. Model Diagnostics & Reduction: Identifying
high-leverage points, outliers, and influential
observations (Cook’s Distance), followed by
feature selection to optimize performance.
5. Business Interpretation &
Recommendations: Translating statistical
coefficients into actionable business intelligence.
Section 2: Data Preparation & Exploratory Data
Analysis (EDA)
2.1 Python Environment & Setup
To execute the data mining workflow, we utilize
standard Python scientific libraries: pandas, numpy,
matplotlib, seaborn, and statsmodels.
