You must forecast a simple seasonal time series. Which model type fits?
Why D is the answer
ARIMA
ARIMA is a classic model for time series, and it suits a simple seasonal forecast.
In detail
- ARIMA suits a simple seasonal series.
- Try simple models first.
Use case
An illustrative scenario, not a real incident.
A shop forecasts simple seasonal sales with ARIMA, which is quick to train and easy to explain.
See it in different ways
flowchart TD
A{"Problem"} -->|"Forecast a time series"| B["ARIMA"]
A -->|"Complex patterns in large data"| C["DNN"]
A -->|"Language tasks"| D["LLM"]stateDiagram-v2 [*] --> Task Task --> PrebuiltApi: a common task Task --> BigQueryMl: data in BigQuery, SQL team Task --> AutoML: custom model, no code Task --> CustomTraining: your own code
flowchart TD
Q{"What kind of forecast?"}
Q -->|"Simple and seasonal"| A0["ARIMA"]
Q -->|"Complex patterns"| A1["Deep model"]Key point: ARIMA suits simple time series.
Why the other options are wrong
- A. A large language model
- A language model is not for simple forecasting.
- B. An image generation model
- That generates images.
- C. A diffusion model
- That generates images.