BA 03 - CIA REVIEW - BUSINESS ACUMEN - DATA ANALYTICS

Note on Big Data, Data Analytics, and Time Series Analysis

Page 4: CONTENT

  • Big Data

  • Data Analytics

  • Time Series Analysis

Page 5: BIG DATA

Page 6: BIG DATA

  • Definition: Big Data refers to vast amounts of data collected from various sources.

  • Subjectivity: The concept of Big Data is subjective and varies based on the size and complexity of an organization.

Page 7: DATA LIFE CYCLE

  • Components:

    • Discover: Involves preparation, exploration, and modeling.

    • Deploy: Focuses on implementation, action, evaluation, AND DELETION.

Page 8: DATA LIFE CYCLE

  • Detailed Orientation:

    • Data: Collection and generation for processing.

    • Analytics: Cleansing, normalizing, aggregating, extracting, and analyzing data.

    • Results: Improving decisions and outcomes for new benefits.

    • Decision: Determining the best course of action to solve problems or answer queries.

Page 9: DATA LIFE CYCLE

  • Quality of Data:

    Big data does not necessarily mean good data.

    • Good Data: Leads to good outcomes, results, and decisions.

    • Bad Data: Results in poor outcomes, results, and decisions.

Page 10: DATA OWNERS AND STEWARDS

  • Data Owner: Responsible for safeguarding data, classifying it, and defining access rules.

  • Data Steward: Manages specific data resources, standardizes data assets across the organization.

Page 11: WHY USE BIG DATA?

  • Opportunities: Combines internal and external data for new opportunities.

  • Decision-Making: Facilitates big decisions, new insights, strategies, and actions.

  • Strategic Asset: Considered a new data asset and strategic asset.

  • Consulting Services: Enhances assurance procedures and consulting services.

Page 12: SOURCES OF BIG DATA

  • Structured Data:

    • Internal documents like sales orders, invoices, purchase requests, and operating expenses.

Page 13: SOURCES OF BIG DATA

  • Unstructured Data:

    • External documents or publicly available documents in human languages, audio, and video from online sources, search engines, and government websites.

Page 14: CHARACTERISTICS OF BIG DATA

  • 7 Vs of Big Data:

    1. Volume: Amount of data created.

    2. Variety: Different forms and formats of data.

    3. Velocity: Speed of data generation.

    4. Veracity: Accuracy and content verification.

    5. Variability: Constantly changing nature of data.

Page 15: CHARACTERISTICS OF BIG DATA

7 Vs of Big Data (cont.):

  • 6. Visualization: Presenting data in understandable graphics and charts.

  • 7. Value: Benefits to organizations, societies, and consumers.

Page 16: RISKS IN BIG DATA

  • Potential Issues:

    • Wrong data leading to poor decisions.

    • Analysis paralysis from excessive data. “Data analytics rich but information poor”

    • Invalid data patterns wasting resources.

    • Poor data governance standards.

    • Lack of data security.

Page 17: DATA QUALITY STANDARDS

R-A-C-T-A-I-C

  1. Relevance

  2. Accuracy

  3. Credibility

  4. Timeliness

  5. Accessibility

  6. Interpretability

  7. Coherence

Page 18: DATA ANALYTICS

Page 19: DATA ANALYTICS

  • Definition: Involves applying quantitative and qualitative tools to big data for insights and opportunities.

Page 20: TYPES OF DATA ANALYTICS

  1. Predictive Analytics

  2. Descriptive Analytics

  3. Prescriptive Analytics

  4. Cognitive Analytics

Page 21: PREDICTIVE ANALYTICS

  • Purpose: Estimates future outcomes based on past and current data.

Page 22: DESCRIPTIVE ANALYTICS

  • Purpose: Describes what has already happened through content and context analysis.

Page 23: PRESCRIPTIVE ANALYTICS

  • Purpose: Assists management in deciding the best course of action among various options.

Page 24: COGNITIVE ANALYTICS

  • Definition: Utilizes AI technologies like machine learning and natural language processing for insights in business transactions.

Page 25: TIME SERIES AND FORECASTING METHODS

The application of Big Data

Page 26: FORECASTING METHODS

  • Classification:

    • Qualitative Methods: Use expert judgment when historical data is unavailable.

    • Quantitative Methods: Use past information to forecast future values based on established patterns; can be quantified; can be assumed that past is prologue.

Page 27: TIME SERIES

  • Definition: A sequence of observations on a variable measured at successive time points.

  • Importance: Understanding past behavior to guide future forecasting methods.

Page 28: TIME SERIES PLOT

  • Purpose: Graphically presents the relationship between time and the variable, aiding in pattern identification.

Page 29: HORIZONTAL PATTERNS

  • Definition: Data fluctuates randomly around a constant mean.

  • Stationary Time Series: Exhibits constant mean and variability over time.

Page 31: TREND PATTERNS

  • Definition: Long-term factors cause a trend in data, influenced by demographic changes, technology, and consumer preferences. specific pattern/constant/smoothing/consistent

  • Derived from horizontal pattern.

Page 33: SEASONAL PATTERNS

  • Definition: Recurring patterns observed over time, such as seasonal sales fluctuations in retail. over a one-year period due to seasonal influences.

Page 34: CYCLICAL PATTERNS

  • Definition: Alternating sequences of data points above and below a trendline, often related to multiyear business cycles.

  • Trend-cycle effects - when cyclical effects are combined with long-term trend effects.

Page 36: IDENTIFYING TIME SERIES PATTERNS

  • Importance: Recognizing patterns in time series data is crucial for selecting.

  • A time series plot should be one of the first analytic tools employed when trying to determine which forecasting method to use.