Time Series Analysis
Analyze time series data, identify patterns, and develop forecasting models
Prompt template
You are an expert in time series analysis and statistical forecasting. Perform a comprehensive analysis of the following time series data. Data description: [DATA – e.g., daily website visitors over the last 2 years, ~730 data points, including holiday information] Forecasting goal: [GOAL – e.g., predict visitor numbers for the next 90 days for capacity planning] Known influencing factors: [FACTORS – e.g., seasonality, marketing campaigns, public holidays, external events] Analysis tool: [TOOL – e.g., Python/statsmodels/Prophet, R/forecast, MATLAB] Please provide: 1. Exploratory analysis (identify trend, seasonality, cyclicity, irregularities) 2. Stationarity test (ADF test) and differencing of the time series if needed 3. Model selection and comparison (e.g., ARIMA, SARIMA, Prophet, Holt-Winters) 4. Complete code for model training, validation, and forecasting 5. Confidence intervals for the forecast and interpretation of uncertainty 6. Anomaly detection for historical outliers
Replace placeholders in [square brackets]: [TOOL – e.g., Python/statsmodels/Prophet, R/forecast, MATLAB]
How to use this template
Copy the prompt, replace the placeholders in square brackets with your details and paste it into the AI assistant of your choice. The more precise your details, the better the result.
Tips for better results
- Ensure your time series has no gaps — missing timestamps should be filled in beforehand.
- Document external events like campaigns or system outages as covariates for better forecasts.
- Split the data into training and test sets and report error metrics such as MAE, RMSE, and MAPE.
Suitable open models
The selection names current open models as a starting point for your own tests. It is not a measurement for this template.
Keywords: #time-series #forecasting #arima #prophet