{% extends 'common/base_v2.html' %} {% block summary %}

Executive Summary

Best Model
{{ best_estimator_label|default('N/A') }}
BIC
{{ best_bic|default('N/A') }}
R² adj
{{ best_rsq_adj|default('N/A') }}
Status
{{ best_classification|default('N/A') }}
Models Tested
{{ n_models_tested|default(0) }}
Combinations
{{ n_combinations_tested|default(0) }}
{% if datamining_warning %}
Data Mining Warning: {{ n_combinations_tested }} model/variable combinations were tested. Results should be validated on out-of-sample data. Statistical significance may be overstated due to multiple testing.
{% endif %}
{% endblock %} {% block content %}

Top {{ ranking_rows|length }} Models

Click column headers to sort. Ranked by composite score (criterion + tests + sign constraints + parsimony).

{% for row in ranking_rows %} {% endfor %}
# ▲ Model ▲ Formula ▲ BIC ▲ AIC ▲ R² adj ▲ Status ▲ Vars ▲ Score ▲
{{ row.rank }} {{ row.model_type }} {{ row.formula }} {{ "%.2f"|format(row.bic) }} {{ "%.2f"|format(row.aic) }} {{ "%.4f"|format(row.rsq_adj) if row.rsq_adj is not none else 'N/A' }} {{ row.classification }} {{ row.n_vars }} {{ "%.3f"|format(row.score) }}

Best Model Details

{% if best_model_info %}
Estimator
{{ best_estimator_label }}
R²
{{ best_model_info.rsq|default('N/A') }}
BIC
{{ best_model_info.bic|default('N/A') }}
AIC
{{ best_model_info.aic|default('N/A') }}
RMSE
{{ best_model_info.rmse|default('N/A') }}

Formula: {{ best_formula }}

Standard Errors: {{ best_cov_type }} {% if se_justification %} — {{ se_justification }} {% endif %}
{% if coefficients %}

Coefficients

{% for coef in coefficients %} {% endfor %}
Parameter Estimate Std. Error t-stat p-value Sig.
{{ coef.name }} {{ "%.4f"|format(coef.estimate) }} {{ "%.4f"|format(coef.std_error) }} {{ "%.3f"|format(coef.tstat) }} {{ coef.pvalue|pvalue_format }} {{ coef.pvalue|significance_stars }}

Significance: *** p<0.001, ** p<0.01, * p<0.05, . p<0.1

{% endif %} {% if best_diagnostics %}

Diagnostics

    {% for diag_name, diag_val in best_diagnostics.items() %}
  • {{ diag_name }}: {{ diag_val }}
    • {% endfor %}
{% endif %} {% else %}
No valid model found. All models were classified as INVALID.
{% endif %}

Specification Test Matrix

Models × tests heatmap. Green = pass, red = fail, gray = not applicable.

{% if test_matrix %}
{% for test_name in test_names %} {% endfor %} {% for model_name, tests in test_matrix.items() %} {% for test_name in test_names %} {% set result = tests.get(test_name, 'n/a') %} {% endfor %} {% endfor %}
Model{{ test_name }}Classification
{{ model_name }} {{ '✓' if result == 'pass' else ('✗' if result == 'fail' else '—') }} {{ tests._classification }}
{% else %}
No test results available.
{% endif %}

Variable Selection

{% if bic_chart_b64 %}

BIC Path

Information criterion at each forward step, by model type.

BIC Path Chart
{% endif %} {% for model_type, sel in variable_selection.items() %}

{{ model_type }}

Selected ({{ sel.selected_vars|length }}): {% for v in sel.selected_vars %} {{ v }}{{ ', ' if not loop.last else '' }} {% endfor %} {% if not sel.selected_vars %} None {% endif %}

{% if sel.rejected_vars %}

Rejected:

    {% for var, reason in sel.rejected_vars.items() %}
  • {{ var }} — {{ reason }}
    • {% endfor %}
{% endif %} {% endfor %}

Variable Transformations

{% if transformations %}
{% for t in transformations %} {% endfor %}
Original Variable Transformation Result Name
{{ t.source }} {{ t.type }} {{ t.name }}
{% else %}

No transformations were applied.

{% endif %}

Transparency & Data Mining

Combinations Tested
{{ n_combinations_tested|default(0) }}
Models Estimated
{{ n_models_tested|default(0) }}
Transformations
{{ transformations|length if transformations else 0 }}
{% if datamining_warning %}
Warning: More than 100 model/variable combinations ({{ n_combinations_tested }}) were evaluated. This increases the risk of spurious results due to multiple testing. Report these numbers in your research for transparency.
{% endif %}
Disclaimer: AutoExperiment automates model selection using econometric criteria and diagnostic tests. However, automated selection does not replace theoretical justification. Results should be interpreted with caution, validated on held-out data when possible, and should not be treated as evidence of causal relationships without further identification arguments.
{% endblock %} {% block extra_js %} {% endblock %}