Which approach is most suitable for making data-driven predictions?

Question

Options

• ABasic SQL queries
• BUsing built-in SQL functions for statistical analysis
• CRelying on external data sources only
• DIgnoring historical data trends

• A
  2% (1)
• B
  90% (46)
• C
  6% (3)
• D
  2% (1)

Explanation

Using built-in SQL functions for statistical analysis (B) is correct because these functions - such as AVG, STDDEV, CORR, PERCENTILE, and window functions like LAG/LEAD - are purpose-built for computing statistical patterns directly from your data, enabling trend analysis and predictive insights without leaving the database.

Why the distractors fail:

• A (Basic SQL queries): Simple SELECT/WHERE/JOIN queries retrieve and filter data but have no predictive capability on their own - they describe, not forecast.
• C (External data sources only): Ignoring your own historical data entirely is counterproductive; external data may supplement predictions but cannot replace internal patterns.
• D (Ignoring historical trends): Predictions depend on historical patterns by definition - discarding them eliminates the very foundation of any data-driven forecast.

Memory tip: Think "Statistics Stays in SQL" - when the question involves prediction or analysis, look for the answer that keeps the analytical power inside the database using built-in statistical functions, not the one that merely queries or avoids data altogether.

No community discussion yet for this question.