Official Data
Official Data Observatory Data structures, trends, and evidence
Load official economic indicators from Ceteris Lab's server export endpoint, plot a time series, calculate growth rates, and discuss limitations.
official data time series plot growth rates limitations
Module 0
Loading CSV Data in Python Loading CSV data in Python
Load wage_sample.csv, inspect rows, and identify variables in a DataFrame.
pandas CSV loading DataFrame preview
Module 0
Descriptive Statistics in Python Descriptive statistics in Python
Calculate means, medians, standard deviations, and grouped wage summaries.
summary statistics groupby interpretation
Module 0
Creating Simple Graphs in Python Creating simple graphs in Python
Create a scatter plot that compares education and wages.
matplotlib scatter plots visual interpretation
Module 0
Mini Python Practice Lab Mini Python practice lab
Run a full beginner workflow: load, inspect, summarize, graph, and interpret.
workflow summaries graphs
Module 1
First Python Data Example First Python data example
Use Python to describe the wage and education relationship without overclaiming causality.
correlation grouped means careful interpretation
Module 2
Simple Regression in Python Regression in Python
Estimate wage on education, draw the fitted line, and read the coefficient output with a statsmodels-compatible fallback.
OLS statsmodels-ready fitted line
Module 2
Fitted Values and Residuals Fitted values and residuals
Calculate fitted wages, residuals, squared residuals, and residual plots from the wage regression.
fitted values residuals diagnostic plots
Module 2
OLS and R-squared by Formula Ordinary Least Squares intuition
Compute the OLS slope, intercept, residual sum of squares, and R-squared step by step.
OLS formulas SSR R-squared
Module 2
Module 2 Regression Project Template Practice regression project
A reusable project notebook students can edit to choose y and x, estimate the model, and write an interpretation.
project workflow student template interpretation
Module 3
WAGE1 Multiple Regression Multiple Regression in Python
Estimate log wage on education, experience, and tenure when the WAGE1 file is installed.
multiple regression WAGE1 statsmodels
Module 3
GPA1 Multiple Regression Holding Other Factors Fixed
Estimate college GPA on high-school GPA and ACT when the GPA1 file is installed.
controls GPA1 coefficient interpretation
Module 3
Simple vs Multiple Regression Simple Regression versus Multiple Regression
Compare the education coefficient before and after adding experience and tenure controls.
simple regression controls coefficient comparison
Module 3
Partialling-Out Demonstration The Partialling-Out Interpretation
Recover a multiple-regression coefficient by using the leftover part of education after controls.
partialling out residuals ceteris paribus
Module 3
Omitted-Variable Bias Example Omitted-Variable Bias
Compare a short wage model with a controlled model and discuss the direction of omitted-variable bias.
omitted-variable bias controls interpretation
Module 3
VIF Calculation Multicollinearity and VIF
Calculate variance inflation factors for education, experience, and tenure.
VIF multicollinearity precision
Module 3
Residuals and Fitted Values Fitted Values and Residuals
Create fitted values and residuals from a multiple-regression wage model.
fitted values residuals diagnostics
Module 3
R-squared Comparison Goodness of Fit in Multiple Regression
Compare R-squared across nested models and separate fit from causal interpretation.
R-squared model comparison fit
Module 4
T Tests with WAGE1 Testing a Single Coefficient Against Zero
Estimate a wage equation and test education, experience, and tenure with real t statistics.
t tests WAGE1 statsmodels
Module 4
One-Sided and Two-Sided Tests One-Sided and Two-Sided Tests
Compare tail choices and hypothesis wording using GPA1.
one-sided tests two-sided tests p-values
Module 4
p-Values and Critical Values p-Values for t Tests
Use scipy to compute p-values and critical values from the t distribution.
p-values critical values scipy
Module 4
Confidence Intervals Confidence Intervals for Regression Coefficients
Build confidence intervals from estimates, standard errors, and t critical values.
confidence intervals WAGE1 inference
Module 4
Linear Combinations Testing One Linear Combination of Parameters
Test whether two coefficients are equal using WAGE2.
linear combinations coefficient equality t_test
Module 4
F Tests for Exclusion Restrictions F Tests for Exclusion Restrictions
Compare restricted and unrestricted models and compute an F statistic.
F tests restricted model unrestricted model
Module 4
R-Squared F Test R-Squared Form of the F Statistic
Use R-squared values to compute the F statistic for valid exclusion restrictions.
R-squared F statistic nested models
Module 4
General Linear Restrictions General Linear Restrictions
Use statsmodels f_test for restrictions beyond simple variable exclusion.
general restrictions f_test housing data
Module 4
Reporting Regression Results Reporting Regression Results Professionally
Assemble coefficient, uncertainty, p-value, fit, and interpretation into a report table.
reporting regression table professional interpretation
Module 4
Applied Inference Project Applied Inference Project
Run a complete inference workflow and draft a concise empirical report.
applied project inference workflow reporting
Module 5
Consistency Simulation What Consistency Means
Simulate OLS under exogeneity and watch estimates concentrate as n grows.
consistency simulation large samples
Module 5
OLS Inconsistency from Endogeneity Inconsistency and Asymptotic Bias
Show estimates converging to the wrong target when x and u are correlated.
inconsistency endogeneity simulation
Module 5
Omitted-Variable Inconsistency Omitted Variable Inconsistency
Compare true and omitted models as sample size grows.
omitted variables probability limits simulation
Module 5
Asymptotic Normality Large-Sample Inference without Normal Errors
Show coefficient distributions becoming approximately normal under several error distributions.
asymptotic normality CLT simulation
Module 5
Nonnormal Errors and Large-Sample Inference Large-Sample t and F Tests
Use 401K to discuss bounded, nonnormal outcomes and approximate inference.
nonnormality large-sample tests 401K
Module 5
Standard Errors Shrink with n Asymptotic Standard Errors
Estimate growing GPA2 subsamples and compare standard errors.
standard errors sample size GPA2
Module 5
Histograms, Skewness, and Transformations Histograms, Normality, and Transformations
Compare WAGE1 residual histograms for wage and log(wage).
histograms skewness log transformations
Module 5
LM Test with CRIME1 The Lagrange Multiplier Test
Compute the n-R-squared LM statistic and compare it with exclusion-test logic.
LM test auxiliary regression CRIME1
Module 5
Asymptotic Efficiency Simulation Asymptotic Efficiency of OLS
Compare OLS with an alternative consistent estimator in simulation.
asymptotic efficiency simulation estimator comparison
Module 5
Applied Asymptotics Project Module 5 Applied Project
Plan a complete large-sample inference workflow with simulations, diagnostics, and limitations.
applied project asymptotics reporting
Module 6
Scaling and Coefficient Interpretation Why Units Matter in Multiple Regression
Convert slopes across wage units and confirm that fitted relationships do not change when units are handled correctly.
scaling unit conversion coefficient interpretation
Module 6
Standardized Betas Standardized Coefficients
Compare raw and standardized coefficients while avoiding causal importance rankings.
standardized beta comparison interpretation
Module 6
Log Models and Percent Changes Level-Log and Log-Level Models
Practice level-log, log-level, and log-log interpretations with exact percentage conversions.
log models percent changes elasticity
Module 6
Quadratic Turning Points Quadratic Terms and Turning Points
Estimate a quadratic model and calculate turning points and marginal effects.
quadratics turning points marginal effects
Module 6
Interaction Effects Interactions between Continuous Variables
Calculate interaction-based marginal effects at meaningful values.
interactions dummy variables marginal effects
Module 6
Centering Interactions Centering Variables before Interactions
Show how centering changes coefficient meaning while preserving fitted values.
centering interactions collinearity
Module 6
Adjusted R-Squared and Model Comparison Adjusted R-Squared and Model Size
Compare candidate specifications using adjusted R-squared and modeling logic.
adjusted R-squared model comparison controls
Module 6
Bad Controls and Precision Controls Over-Control and Bad Controls
Distinguish harmful controls from safe precision controls.
bad controls precision controls design
Module 6
Prediction and Prediction Intervals Prediction with Multiple Regression
Create fitted predictions and compare mean and individual prediction intervals.
prediction prediction intervals uncertainty
Module 6
Log Prediction and Smearing Predictions when the Dependent Variable Is Logged
Compare naive and smearing-adjusted retransformation from log outcomes.
log prediction smearing retransformation
Module 6
Bootstrap Standard Errors Project Module 6 Applied Forecasting Project
Bootstrap a coefficient and write a careful interpretation with uncertainty.
bootstrap standard errors project
Module 7
Binary Variables Creating Binary Variables
Create and audit binary indicators from transparent rules.
dummy variables qualitative information Python statsmodels
Module 7
Group Mean Comparisons Comparing Two Means with Regression
Show how a dummy-only regression reproduces a two-group mean difference.
dummy variables qualitative information Python statsmodels
Module 7
Dummy Variables in Log Models Dummy Variables in Log-Dependent Models
Convert log-dummy coefficients using approximate and exact percentages.
dummy variables qualitative information Python statsmodels
Module 7
Multiple Categories Multiple Categories
Encode categories with a base group and compare coefficient meanings.
dummy variables qualitative information Python statsmodels
Module 7
Dummy Interactions Binary by Binary Interactions
Interpret binary-by-binary interactions as conditional group differences.
dummy variables qualitative information Python statsmodels
Module 7
Different Slopes and Centering Binary by Continuous Interactions
Estimate group-specific slopes and show how centering changes the reference point.
dummy variables qualitative information Python statsmodels
Module 7
Chow-Style Group Difference Tests Full Regression Differences Across Groups
Use interaction restrictions to test full group differences.
dummy variables qualitative information Python statsmodels
Module 7
Linear Probability Model Linear Probability Model
Estimate an LPM and interpret coefficients as probability-point changes.
dummy variables qualitative information Python statsmodels
Module 7
LPM Robust Standard Errors Limitations of the Linear Probability Model
Inspect fitted probabilities and use HC1 robust standard errors.
dummy variables qualitative information Python statsmodels
Module 7
Policy Evaluation and Self-Selection Policy Evaluation and Self-Selection
Compare randomized and self-selected treatment comparisons.
dummy variables qualitative information Python statsmodels
Module 7
Discrete Dependent Variables Discrete Dependent Variables
Describe count outcomes and explain why discrete models may be needed later.
dummy variables qualitative information Python statsmodels
Module 7
Module 7 Capstone Module 7 Capstone
Complete a full qualitative-variable modeling checklist.
dummy variables qualitative information Python statsmodels
Module 8
Simulating Heteroskedasticity What Is Heteroskedasticity?
Generate and visualize changing variance.
heteroskedasticity robust standard errors Python statsmodels
Module 8
Conventional versus Robust Standard Errors Robust Standard Errors
Compare conventional, HC0, HC1, HC2, and HC3 standard errors.
heteroskedasticity robust standard errors Python statsmodels
Module 8
Breusch-Pagan Test The Breusch-Pagan Test
Run and manually audit the Breusch-Pagan workflow.
heteroskedasticity robust standard errors Python statsmodels
Module 8
White Test The White Test
Use full and fitted-value White diagnostics.
heteroskedasticity robust standard errors Python statsmodels
Module 8
Robust Joint Tests Robust Joint Tests
Test multiple restrictions with robust covariance.
heteroskedasticity robust standard errors Python statsmodels
Module 8
Weighted Least Squares Weighted Least Squares Intuition
Choose inverse-variance weights and compare OLS with WLS.
heteroskedasticity robust standard errors Python statsmodels
Module 8
Population and Group-Size Weights Group Means, Population Weights, and Aggregated Data
Use group size as a precision weight.
heteroskedasticity robust standard errors Python statsmodels
Module 8
Feasible GLS Feasible GLS
Estimate a variance function and fit FGLS.
heteroskedasticity robust standard errors Python statsmodels
Module 8
Robust Standard Errors after WLS What If the WLS Variance Model Is Wrong?
Compare WLS conventional and robust standard errors.
heteroskedasticity robust standard errors Python statsmodels
Module 8
Prediction Intervals with Changing Variance Prediction under Heteroskedasticity
Plot prediction intervals under changing variance.
heteroskedasticity robust standard errors Python statsmodels
Module 8
LPM Robust Inference The Linear Probability Model Revisited
Estimate a linear probability model with HC1 robust standard errors.
heteroskedasticity robust standard errors Python statsmodels
Module 8
Module 8 Capstone Module 8 Capstone
Complete a full diagnostic and correction workflow.
heteroskedasticity robust standard errors Python statsmodels