Here’s a summary of what you have learnt so far in this session:

- First, you began by exploring the different applications of forecasting.
- You then understood that forecasting can be done in two ways:
- Qualitative forecasting
- Quantitative forecasting

- After that, you learnt that time series forecasting comes under quantitative forecasting.
- After getting exposed to the Air Passenger traffic problem statement, you learnt the basic steps for forecasting.
- Define the problem
- Quantity
- Granularity
- Frequency
- Horizon

- Collect the data
- Three important characteristics of time series data:
- Relevant
- Accurate
- Long enough

- Three important characteristics of time series data:
- Analyse the data
- You analysed the Air Passenger traffic dataset and found that the dataset has missing values.
- Then, you learnt the various methods to handle the missing values and imputed the missing values in the Air Passenger traffic dataset.
- You also explored some of the methods to detect outliers and understood that Air Passenger traffic dataset does not have any outliers.
- You also explored the methods to decompose the time series data into its components.
- Additive Seasonal Decomposition
- Multiplicative Seasonal Decomposition

- We will now learn how to build and evaluate the forecast model. (Upcoming Sessions)

- Define the problem