Forecasting
Time-Series Forecasting with Seasonal Models
Model seasonality, compare smoothing approaches, and communicate forecast intervals for operations planning.
Participants work with tourism, energy, and logistics series that exhibit Thai holiday effects. Sessions contrast exponential smoothing, ARIMA baselines, and lightweight machine benchmarks while keeping emphasis on interval honesty and revision policies when new observations arrive.
- Format
- Blended cohort
- Duration
- 6 weeks · blended
- Level
- Intermediate
- Software
- R, Python
- Topic
- Forecasting
- Tuition
- 9,500 THB · informational until admissions confirms
What is included
- Seasonal adjustment walkthroughs with transparent parameter notes
- Hands-on labs comparing ARIMA vs. ETS on the same holdout window
- Forecast communication templates for executives
- Office hours on backcasting when data start late
- Capstone presentation with recorded critique
Outcomes you can evidence
- Select a baseline model with explicit trade-offs stated upfront
- Explain forecast revisions without overclaiming precision
- Document data frequency assumptions for downstream teams
Lead contact
Marcus Yeung
Forecasting lead for infrastructure demand studies; former central bank trainee instructor.
Participant voices
Interval narration finally clicked after the seasonal models lab — I still quote the template in our ops review.
Questions we expect
Ready to talk fit?
Request information Admissions responds with prerequisites and employer letter options.