Exclusive 40% Discount: Time Series Bootcamp – Forecasting with ML, Deep Learning & Foundation Models (Code: HUI40)
Special 40% Discount for the IBridgePy Community
Save a massive 40% using promo coupon code HUI40 (Save up to $200!).
How do classical statistical forecasting methods stack up against modern gradient-boosted trees, deep learning architectures, and pre-trained time-series foundation models like Chronos and TabPFN?
In this comprehensive, 2-day hands-on bootcamp hosted by Packt, three veteran forecasting practitioners — Graeme Davidson, Lei Ma, and Jeffrey Tackes — guide you through benchmarking, selecting, and deploying production time-series models across real-world datasets.
We are thrilled that IBridgePy has negotiated an exclusive 40% discount for our community with coupon code HUI40 — saving you up to $200 on full 2-day bootcamp access!
📅 Bootcamp Snapshot
Saturday, Nov 28 – Sunday, Nov 29, 2026
10:00 AM – 2:00 PM EST (3:00 – 7:00 PM UTC) each day
2 Days • 8–10 Hours Total Live Interactive Training
Graeme Davidson, Lei Ma & Jeffrey Tackes
Packt - ML & AI Engineering
40% Off with Code HUI40 ($299.99 vs $499.99)
Exclusive Ticket Pricing Breakdown (Coupon: HUI40)
Save $200 on the 2-Day Time Series Bootcamp
Master state-of-the-art time series forecasting across ML, PyTorch, and foundation models. Discount applied automatically.
Enroll with 40% Discount (Code: HUI40) →Includes 2 full days of live instruction, evaluation harness, GitHub code, and complete recordings.
Why Time Series Forecasting is Critical for IBridgePy Algorithmic Traders
Time series modeling is the mathematical heartbeat of algorithmic trading. Whether estimating price trajectories, forecasting volatility cones, predicting trading volume distributions, or managing multi-asset rebalancing:
- Eliminating Lookahead Bias: In financial markets, accidentally training models on future data invalidates your backtest. The bootcamp teaches time-aware cross-validation and strict temporal split methodologies.
- Probabilistic Forecasts & Risk Bounds: Single-point forecasts fail when market volatility expands. Generating prediction intervals and quantile forecasts allows IBridgePy trading algorithms to size positions dynamically based on forecast uncertainty.
- Foundation Models for Trading: Discover how zero-shot and few-shot time-series foundation models (like Chronos-2 and TabPFN-TS) compare against fine-tuned gradient-boosted trees when deployed on real sequential data.
What You Will Master Over the 2 Days
- Modern Time-Series Stack: Hands-on experience with StatsForecast, MLForecast, NeuralForecast, TimeCopilot, Chronos-2, and TabPFN-TS.
- Feature Engineering & Exogenous Variables: Construct calendar lags, rolling window statistics, Fourier terms, and external covariates without data leakage.
- Hierarchical Reconciliation: Reconcile bottom-up individual asset predictions with top-down index/portfolio level forecasts.
- Backtesting & Leaderboard Evaluation: Assess forecasting models against shared leaderboards balancing computational cost, forecast stability, and business accuracy.
- Ensembling & Model Selection: Learn when to deploy lightweight statistical models, when gradient boosting wins, and when neural foundation models excel.
Meet Your Instructors
- Graeme Davidson: Lead Data Scientist at Retail Express and co-author of Time Series with PyTorch. Expert in connecting forecasting models with high-volume operational decisions.
- Lei Ma: Physicist-turned-data scientist and co-author of Time Series with PyTorch. Specializes in deep neural forecasting architectures and probabilistic uncertainty modeling.
- Jeffrey Tackes: Founder of Forecast Academy and co-author of Modern Time Series Forecasting with Python (2nd Edition). Focuses on practical backtesting, model evaluation, and demand planning.
Frequently Asked Questions
The 40% discount is applied automatically when using our link. Coupon code HUI40 takes $200 off the full bootcamp price instantly at checkout on Luma.
Individual Day-1 and Day-2 passes are available (each discounted by 40% to $167.99 with code HUI40). Full pass holders get complete recordings of both days.
All labs are set up for Google Colab using Python, StatsForecast, NeuralForecast, LightGBM, and open-source foundation model libraries.
Master the Future of Time Series Forecasting
Take your forecasting models from classical baselines to state-of-the-art foundation models. Save 40% ($200 off) with code HUI40.
Register on Luma with 40% Off →Coupon HUI40 applied automatically • 2 Days of Live Training • Full Recordings & Code Included