All examples for Regression Forecasting Assistant

Monthly Sales Trend Forecasting

Linear regression on 12 months of steadily growing sales data to forecast next quarter. Demonstrates high R² (99.9%), validated assumptions, and extrapolation warnings with confidence intervals.

Input

What you provide to the skill

I have 12 months of monthly sales data (Jan: 50K, Feb: 52K, Mar: 55K, Apr: 58K, May: 61K, Jun: 64K, Jul: 67K, Aug: 70K, Sep: 72K, Oct: 75K, Nov: 78K, Dec: 81K). Forecast next 3 months.

Output

What the skill generates for you

============================================================
REGRESSION ANALYSIS RESULTS

DATA QUALITY

  • Observations: 12 months
  • Missing values: 0
  • Sales range: $50K - $81K
  • Mean sales: $65.2K
  • Std deviation: $10.28K
  • Correlation coefficient: 0.9996
  • Sample size check: PASS (n=12)
  • Variance check: PASS
  • Quality gates: ALL PASSED

MODEL FIT

  • R² = 0.9991 (explains 99.9% of variance)
  • Adjusted R² = 0.9990
  • F-statistic: 11412.80 (p = 0.000000)
  • Slope: $2.850K per month (p = 0.000000)
  • Intercept: $46.73K

Interpretation: Sales increase by $2.85K per month on average

ASSUMPTION CHECKS

  • Linearity: OK (residuals centered near zero)
  • Homoscedasticity: OK (Breusch-Pagan p = 0.2083)
  • Normality: OK (Shapiro-Wilk p = 0.6911)
  • Independence: OK (Durbin-Watson = 1.619)

2025 FORECASTS (Next 3 Months)

January 2025: $83.8K
95% Confidence Interval: $83.3K - $84.2K
95% Prediction Interval: $82.9K - $84.6K
February 2025: $86.6K
95% Confidence Interval: $86.1K - $87.1K
95% Prediction Interval: $85.8K - $87.5K
March 2025: $89.5K
95% Confidence Interval: $88.9K - $90.0K
95% Prediction Interval: $88.6K - $90.4K

EXTRAPOLATION WARNING

Forecasting 3 months beyond training data (months 1-12).
Predictions assume linear trend continues unchanged.
Uncertainty increases with forecast horizon.

BUSINESS RECOMMENDATIONS

  1. Model shows strong linear growth trend (R² = 99.91%)
  2. Expected monthly increase: $2.85K (very consistent)
  3. Use forecasts for planning with +/-15% contingency buffer
  4. Update model monthly as new data arrives to improve accuracy
  5. CORRELATION ≠ CAUSATION - model shows trend, not underlying drivers
  6. Consider external factors: seasonality, market conditions, competition
  7. Model assumptions are satisfied - forecasts are statistically valid
  8. Predicted Q1 2025 total: $259.9K
View all examples for Regression Forecasting Assistant

About This Skill

Build validated regression models for business forecasting with automatic assumption checking, diagnostic plots, confidence intervals, and extrapolation warnings.

View Skill Details

More Examples

Marketing Spend vs Conversion Analysis

Regression analysis on noisy marketing data showing moderate fit (R²=47%). Demonstrates honest model quality assessment, interpolation within training range, and actionable recommendations for budget planning.

Small Sample Demand Forecasting with Caveats

Regression on only 8 weeks of demand data, explicitly addressing small sample concerns. Demonstrates appropriate warnings, wider confidence intervals, AIC model comparison, and conservative recommendations.

View all examples