Extract Consolidation Metrics from Employment Transition Analysis

Phase 2 Optimized Function Calculates detailed consolidation metrics by comparing original pipeline_result with consolidated result from analyze_employment_transitions(). Provides insights into how many persons and contracts are affected by consolidation across all consolidation modes (temporal, employer, both, none).

Now includes intelligent consolidation data generation when actual consolidated data is not available, plus single-pass aggregation patterns for maximum efficiency.

Usage

extract_consolidation_metrics(
  original_data,
  consolidated_data = NULL,
  consolidation_mode = "none",
  employer_var = NULL,
  min_lag = 8,
  consolidation_type = "both"
)

Arguments

original_data

Original pipeline_result data.table from vecshift output

consolidated_data

Consolidated data.table after consolidation (if any)

consolidation_mode

Character string specifying consolidation mode used

employer_var

Character string specifying employer variable (if applicable)

min_lag

Numeric gap threshold for employer consolidation (if applicable)

consolidation_type

Character string specifying temporal consolidation type (if applicable)

Value

List containing consolidation metrics:

• consolidation_summary: Overall consolidation statistics

• person_level: Person-level consolidation metrics

• distribution: Distribution of consolidation patterns

• employer_specific: Employer-specific metrics (when applicable), including:

  • unique_employers: Vector of unique employer IDs

  • n_unique_employers: Count of unique employers

  • employer_details: Per-employer consolidation metrics data.table

  • employer_summary: Aggregated employer statistics

  • employer_consolidation: Person-employer combination metrics

  • employer_person_stats: Person-employer statistics

• temporal_specific: Temporal consolidation metrics (when applicable)

Details

The function analyzes consolidation impact across multiple dimensions:

• Overall Impact: Total contracts vs consolidated periods

• Person-Level: Individual consolidation patterns and averages

• Distribution Analysis: How many contracts consolidated into single periods

• Employer Analysis: Consolidation patterns by employer (when applicable)

Handles all consolidation modes:

• none: Returns baseline metrics (no consolidation)

• temporal: Analyzes over_id-based consolidation

• employer: Analyzes same-employer consolidation

• both: Analyzes sequential consolidation effects

Examples

if (FALSE) { # \dontrun{
# Load sample data and run transition analysis
result <- readRDS("data/sample.rds")

# Run with temporal consolidation
transitions <- analyze_employment_transitions(
  pipeline_result = result,
  consolidation_mode = "temporal",
  consolidation_type = "both"
)

# Extract consolidation metrics
metrics <- extract_consolidation_metrics(
  original_data = result,
  consolidated_data = transitions_data_used_internally,  # Would need to be captured
  consolidation_mode = "temporal",
  consolidation_type = "both"
)

# View overall impact
print(metrics$consolidation_summary)
print(metrics$person_level)
} # }