Predictive Data Analytics

Predictive Data Analytics

Predictive Data Analytics

  • Respond to the following five (5) questions related to one of the learning objectives covered in this course.
  • For question 2, confirm your answers with examples of data sets and/or visualizations.
  • While you may choose these from the sample data sets provided in the resources listed for this course, It is strongly recommended that you search for new data sources to use as examples.

Questions:

  • Differentiate between various types (Descriptive, Predictive, or Prescriptive) of data an organization may use to assess organizational performance.
    • Provide an example for each data source.
    • Highlight the purpose of the data sources, the metric(s) it explains, and what kind of decision it would help justify.
  • Create a data visualization graphic that incorporates appropriate data sets for one of the three types.
    • Consider one of the data sets you have shared in question number 1 of this workbook.
  • Evaluate the benefits of at least two different data analysis methods.
    • Share an example of each.
    • Explain how, when, and why these methods have been used in a business situation.
  • Justify a strategic choice based on a data analysis method.
    • Use the data analysis method in Week 3 or another example of your choice.
  • Assess how big data can influence organizational performance.
    • You may consider using an example if you find that helpful to support your argument.
    • Consider how data can create insight into a business problem and provide a sense of decision-making justification.

4 pages

at least 1 source plus book  Jaggia, S. (2023). Business analytics: Communicating with numbers (2nd ed.). McGraw-Hill Higher Education.

Predictive Data Analytics

Check our essay writing services here

APA

Predictive Data Analytics

1) Descriptive vs. Predictive vs. Prescriptive data for organizational performance

Descriptive data summarizes what happened and how performance looked over a period. Examples: monthly revenue, customer satisfaction (CSAT), Net Promoter Score (NPS), defect rates, on-time delivery, or retail sales by channel. Purpose: establish baselines, detect trends/seasonality, and monitor KPIs. Example data source: U.S. Census Monthly Retail Trade time series (sales levels by month). Metrics explained: total sales level, month-over-month and year-over-year deltas, seasonality. Decision justified: staffing/stocking adjustments or budget phasing (“increase Q4 retail staffing by 10% given seasonal lift observed each year”). (Jaggia, 2023). Census.gov

Predictive data estimates what is likely to happen, typically using models trained on historical features. Examples: churn likelihood by customer, next-period sales, lead conversion probability, late-invoice risk. Purpose: forecast outcomes and prioritize interventions. Example dataset: IBM Telco Customer Churn (features like contract type, tenure, charges) used to predict whether a subscriber will churn next month. Metrics explained: predicted probability of churn, lift/ROC-AUC, forecasted attrition volume. Decision justified: proactive retention offers to high-risk, high-CLV customers; queue sizing for retention teams. (Jaggia, 2023). IBM+1

Prescriptive data recommends what to do to achieve a goal under constraints (budget, capacity, SLAs). It uses optimization/simulation layered on descriptive/predictive inputs. Examples: route optimization for delivery fleets, price-promotion optimization, inventory reorder policies. Purpose: choose the…

Tags