Chapter 1 Literature Review Order Forecasting Methods

Seasonal variations method

Seasonal variations are repetitive and predictable patterns that occur within a given time series. Examples of seasonal patterns include increased sales during holiday seasons, fluctuations in temperature throughout the year, or variations in stock market performance during different months.
Seasonality is typically a key factor in projecting fundamental business decisions, including those related to production, inventory management, people, marketing, and finance. A pattern of change that recurs year after year is referred to as seasonal. If the data is monthly, it might be every month, every three, four, or six months. Usually, every quarter for data is reported quarterly.
In the seasonal variation method, how much demand will be in each season for next year is predicted as follows.

1. 
   In a seasonal variation, the seasonal average of the same months is calculated for each cycle every 3 years (each year in this case).
   
2. 
   The total annual average demand is then calculated.
   
3. 
   The seasonal index is calculated by dividing the average monthly demand by the annual average demand.
   
4. 
   We estimate the total annual demand for next year.
   

To find out how much monthly demand there is for each season of the next year, we divide the estimated annual demand by the number of months in the year and multiply it by the seasonal index.
A season's average comparison to the cycle's mean is indicated by the seasonal index. A seasonal index value can be expressed as a percentage or a decimal. The average cycle mean is considered 100% or the decimal 1.00. This is because seasonal indices are computed as average percentages; therefore, averaging these percentages will always result in the real average, or mean, which is denoted as 100%. In order to maintain the same connection while working with decimal values for seasonal indices, we convert to the decimal 1.00 rather than 100%.
There are benefits to taking into account options for the most popular seasonal index processes when an accurate seasonal index is required for commercial applications. Currently, trend and seasonal components of the time series are successively separated to create seasonal indices when they are required for commercial applications. In some applications, a seasonal index's accuracy might be crucial.
It's important to note that the seasonal variations method assumes that the seasonal pattern is repetitive and remains relatively consistent over time. However, if the seasonal pattern changes significantly, alternative approaches may be required to capture and adjust for these changes.
Overall, the seasonal variations method is valuable for understanding and analyzing time series data by separating the seasonal component from other underlying patterns, ultimately leading to more accurate insights and predictions.

3.