Chapter 8 : Perform the Analysis -Predictive Analytics

Perform the Analysis: Predictive Analytics

The AMPS Model

  • AMPS Cycle continues with Performing the Analysis.

  • Focus on predictive analytics.

Defining Predictive Analytics (LO 8-1)

  • Definition of Predictive Analytics:

    • Predictive analytics addresses the question of whether something or some event will happen in the future.

    • Especially useful when there is a strong correlation between the past and future.

    • Defined as analytics performed to provide foresight by identifying patterns in historical data, judging likelihood or probability.

Deterministic vs. Probabilistic Analyses

  • Deterministic Analyses:

    • More factual, reporting known facts (covered in Chapters 6 and 7).

    • All events known beforehand.

  • Probabilistic Analyses:

    • Include predictive and prescriptive analytics that provide less deterministic output.

    • Judging likelihood and probability of future events or outcomes.

    • Element of chance is present.

Examples of Probabilistic Models

  1. Customer Loan Repayment Prediction:

    • Helps decide whether to extend credit to potential borrowers.

  2. Financial Statement Restatement Prediction:

    • Leads to projecting if a company may need to restate its financial statements.

  3. Financial Statement Fraud Prediction:

    • Identifying potential fraudulent companies.

  4. Future Performance Prediction:

    • Includes predicting sales, net income, and cash flows.

Types of Data Analytics

  • Classification of various types presented in an exhibit (not included here).

Classification (LO 8-2)

  • Classification Defined:

    • A predictive analytics technique used to separate or classify a sample (or population) into two or more groups or classes.

    • Aims to predict a probabilistic outcome based on forecasts.

  • Classification Groups or Classes Examples:

    • Bankrupt/Not Bankrupt

    • Fraud/No Fraud

    • Misstatement/No Misstatement

    • Audit Client Acceptance/Client Rejection

    • Extend Loan/Do Not Extend Loan

    • Continue as Going Concern/Do Not Continue as Going Concern

Bankruptcy Classification: Altman’s Z

  • Factors that Predict Bankruptcy:

    1. X1: Working Capital / Total Assets

    • Measures liquidity level in relation to the size of the company.

    1. X2: Retained Earnings / Total Assets

    • Measures long-term profitability.

    1. X3: Earnings Before Interest and Taxes / Total Assets

    • Measures short-term profitability.

    1. X4: Market Value of Stockholders’ Equity / Book Value of Total Debt

    • Measures long-term solvency.

    1. X5: Sales / Total Assets

    • Measures asset efficiency.

Bankruptcy Classification: Decision Rules

  • Original Z-score Formula:
    Z=1.2X1+1.4X2+3.3X3+0.6X4+1.0X5Z = 1.2X1 + 1.4X2 + 3.3X3 + 0.6X4 + 1.0X5

  • Decision Rules:

    • If Z < 1.80: Classify as significant risk of bankruptcy ("distress zone").

    • If 1.80 ext{ ≤ } Z < 3.00: Classify as at risk of bankruptcy ("gray zone").

    • If Z≥3.00Z ≥ 3.00: Classify as not currently at risk of bankruptcy ("safe zone").

Bankruptcy Classification: Base Analysis

  • Bankruptcy Zone Based on Altman’s Z:

    • Significant risk of bankruptcy ("distress zone"): 305 Firms.

    • At risk of bankruptcy ("gray zone"): 516 Firms.

    • Nonbankrupt ("safe zone"): 1,508 Firms.

    • Total firms included in analysis: 2,329.

    • Results from the retail industry analysis (2009-2017).

Loan Extension Classification

  • Criteria Evaluated by Banks and Lenders:

    • Creditworthiness based on borrower characteristics such as:

    • Credit scores

    • Work history

    • Total debt levels

    • Debt-to-income ratios

    • Home ownership status

Loan Extension Classification: Examples

  • Amount of Requested Loan:

    • Greater loan requests likely lead to less acceptance; expected to have a negative (-) relationship with loan acceptance.

  • Employment Length:

    • Longer employment length likely leads to greater acceptance due to job stability; expected to have a positive (+) relationship.

  • Debt-to-Income Ratio:

    • Higher ratios likely lead to lesser acceptance; expected to have a negative (-) relationship.

Fraud/No Fraud Classification

  • Beneish Study (1999):

    • Analyzed companies from 1982-1992 to predict fraud based on factors outlined in SEC releases.

Beneish’s Factors for Predicting Fraud

  • **M-score Factors:

    1. Increase in receivables compared to previous period.

    2. Decline in gross margin percentage.

    3. Decline in asset quality index.

    4. Sales growth increase.

    5. Decrease in depreciation expense.

    6. Decrease in selling, general and administrative costs.

    7. Increase in leverage (debt).

    8. Higher accruals to total assets ratio.