Threshold Analysis Report

Analysis Summary

This analysis was performed on {{ dataset_name }} to find optimal thresholds for predicting events based on leading indicators.

We analyzed {{ total_records }} records, looking at the relationship between {{ score_column }} and {{ target_column }}.

{% if category_column %}

The data was also broken down by {{ category_column }} to understand patterns across different categories.

{% endif %}

Analysis completed on {{ analysis_date }}.

Key Findings

{% for result in top_results %}

{% if loop.index == 1 %}Warning Level{% elif loop.index == 2 %}High Risk Level{% else %}Critical Level{% endif %}

{{ score_column }} ≥ {{ "%.2f"|format(result.threshold) }}

Finding Probability: {{ "%.1f%%"|format(result.event_probability) }}
Captures {{ "%.1f%%"|format(result.capture_rate) }} of all findings
Based on {{ result.n_observations }} observations

{% if loop.index == 1 %} Action: Monitor closely and investigate if needed. {% elif loop.index == 2 %} Action: Investigate and take preventive measures. {% else %} Action: Immediate investigation required. {% endif %}

{% endfor %}

Threshold Validation

The selected thresholds were validated against historical data:

Threshold {{ "%.2f"|format(result.threshold) }} captures {{ "%.1f%%"|format(result.capture_rate) }} of all findings

{% if category_metrics %}

Category Analysis

Here's how different {{ category_column }}s behave in terms of risk and events:

{% for metric in category_metrics %} {% endfor %}

{{ category_column }}	Number of Records	Number of Events	Event Rate	Average Score
{{ metric.category }}	{{ metric.record_count }}	{{ metric.event_count }}	{{ "%.1f%%"\|format(metric.event_rate * 100) }}	{{ "%.2f"\|format(metric.avg_score) }}

{% endif %}

Visualizations

The following charts show the relationships between scores, events, and categories in your data: