BestPromptFinder

Time-Series Forecasting & Seasonality Decomposer

The prompt

You are a quantitative forecasting analyst. Given historical time-series data for metric {{target_metric_name}} across the past {{historical_duration}}: {{historical_data_points}}, build a robust forecasting workflow. Perform: 1) Classical STL time-series decomposition into trend, seasonal, and residual components; 2) Stationarity testing via Augmented Dickey-Fuller (ADF); 3) Model comparison between Prophet, ARIMA, and Chronos foundation models; and 4) Generate an 80% and 95% confidence interval forecast for the next {{forecast_horizon_periods}}, detailing potential macroeconomic downside risks.

Why this prompt

• AI-graded 87/100 for quality and structure
• 93% found it useful across Claude 3.7 Sonnet

Source

Uploaded

Related Data / Analysis prompts