Business Analytics 101: Introduction to Business Analytics Types of Analytics Descriptive Diagnostic Predictive Prescriptive

What This Is

Business analytics involves using data to inform business decisions. There are four main types of analytics: Descriptive, Diagnostic, Predictive, and Prescriptive. Each type helps answer different questions about business performance and future outcomes. For example, a retail company might use Descriptive analytics to analyze sales data, Diagnostic analytics to identify trends and patterns, Predictive analytics to forecast future sales, and Prescriptive analytics to recommend optimal inventory levels.

Key Formulas & Metrics

• Mean Absolute Error (MAE) = (1/n) Σ|yᵢ − ŷᵢ|: average absolute forecast error, where yᵢ is actual value, ŷᵢ is forecast value, and n is number of observations.
• R² = 1 − (SS_res / SS_tot): proportion of variance explained by the regression model, where SS_res is residual sum of squares and SS_tot is total sum of squares.
• Coefficient of Determination (R²) = 1 - (SSE / SST): proportion of variance explained by the regression model, where SSE is sum of squared errors and SST is total sum of squares.
• Mean Squared Error (MSE) = (1/n) Σ(yᵢ − ŷᵢ)²: average squared forecast error.
• Root Mean Squared Error (RMSE) = √(MSE): square root of average squared forecast error.
• Akaike Information Criterion (AIC) = n * ln(MSE) + 2k: measure of model quality, where n is number of observations, MSE is mean squared error, and k is number of parameters.
• Bayesian Information Criterion (BIC) = n * ln(MSE) + k * ln(n): measure of model quality, where n is number of observations, MSE is mean squared error, and k is number of parameters.
• Confusion Matrix = [TP, FP, FN, TN]: table of true positives, false positives, false negatives, and true negatives.
• Accuracy = (TP + TN) / (TP + FP + FN + TN): proportion of correct predictions.
• Precision = TP / (TP + FP): proportion of true positives among all positive predictions.
• Recall = TP / (TP + FN): proportion of true positives among all actual positives.

Step-by-Step Procedure

1. Define the business problem: Identify the question or opportunity that needs to be addressed.
2. Collect and preprocess data: Gather relevant data, handle missing values, and transform data into a suitable format.
3. Choose the analytics type: Select the type of analytics that best addresses the business problem (Descriptive, Diagnostic, Predictive, or Prescriptive).
4. Develop a model or analysis: Apply statistical or machine learning techniques to analyze the data and answer the business question.
5. Evaluate the model: Assess the model's performance using metrics such as MAE, R², or AIC.
6. Communicate results: Present the findings in a clear and actionable way to stakeholders.

Common Mistakes

• Mistake: Confusing correlation with causation.
• Correction: Establish a clear cause-and-effect relationship between variables before drawing conclusions.
• Mistake: Misinterpreting p-values.
• Correction: Understand that p-values indicate the probability of observing the data (or more extreme) if the null hypothesis is true, not the probability that the null hypothesis is true.
• Mistake: Using the wrong error metric for a business problem.
• Correction: Choose an error metric that aligns with the business objective (e.g., MAE for forecasting, accuracy for classification).

Software / Tool Tips

• Python with pandas/scikit-learn: Use the pandas library for data manipulation and the scikit-learn library for machine learning.
• R: Utilize the dplyr package for data manipulation and the caret package for machine learning.
• Excel: Employ the Analysis ToolPak add-in for statistical analysis and the Power BI tool for data visualization.
• Tableau: Use the Data Analysis feature for statistical analysis and the Storytelling feature for data visualization.

Quick Practice Problem

Problem: Compute the moving average forecast for period 4 given data: 100, 110, 105, 115.

Answer: (100 + 110 + 105 + 115) / 4 = 110.

Explanation: The moving average forecast is the average of the last four periods.

Last-Minute Cram Sheet

• Descriptive analytics: summarizes and describes historical data.
• Diagnostic analytics: identifies trends, patterns, and correlations in data.
• Predictive analytics: forecasts future outcomes based on historical data.
• Prescriptive analytics: recommends optimal actions or decisions based on predictive analytics.
• MAE: average absolute forecast error.
• R²: proportion of variance explained by the regression model.
• AIC: measure of model quality.
• BIC: measure of model quality.
• Confusion matrix: table of true positives, false positives, false negatives, and true negatives.
• Accuracy: proportion of correct predictions.
• Precision: proportion of true positives among all positive predictions.
• Recall: proportion of true positives among all actual positives.
• ⚠️ p-value is NOT the probability that H₀ is true – it’s the probability of observing the data (or more extreme) if H₀ is true.