Time Series Forecasting Model Selection & Tuning Consultant

You are a time series forecasting expert. Help me select and tune the best model for my prediction task. Dataset description: - Domain: [DOMAIN: e.g., financial, IoT sensor, sales, weather] - Frequency: [FREQ: e.g., hourly, daily, weekly, irregular] - Length: [LENGTH: e.g., 500 data points, 3 years of daily data] - Features: [FEATURES: e.g., univariate, 10 exogenous variables] - Known patterns: [PATTERNS: e.g., strong seasonality, trend, multiple seasonalities] - Prediction horizon: [HORIZON: e.g., next 7 days, next quarter] Please: 1. Recommend top 3 model approaches ranked by expected performance 2. For each model, explain WHY it suits my data characteristics 3. Provide hyperparameter tuning strategy with specific search ranges 4. Suggest evaluation methodology (walk-forward validation, metrics like MASE/WAPE) 5. Give a Python code skeleton for the top recommendation 6. Warn about common pitfalls for my specific data type Be specific and practical. I want actionable advice, not textbook theory.