Professional Documents
Culture Documents
• The cyclical effect is the wave –like fluctuation around the trend. The up-down
wave like fluctuations around the trend rarely repeats at fixed intervals of time
and the magnitude of fluctuations also tends to vary. Decomposition methods
(we’ll discuss in later lectures) can be extended to analyze cyclical data.
Forecasting techniques for cyclical data are used whenever the business
cycle influences the variable of interest. Examples are economic markets and
competitive factors.
• Shift in popular tastes occur. Examples are fashion, music and food.
• Shift in population occur. Examples are war, epidemics and natural disasters.
Techniques should be considered when forecasting cyclical series include classical
decomposition, econometric models, multiple regression and ARIMA models etc.
Moving Average method for forecasting :
𝑌𝑡+𝑝 = 𝑎𝑡 + 𝑏𝑡 𝑝
where k = number of periods in the moving average and
p=number of periods ahead to be forecast.
Exercise
• Use double moving average method on the data given
in Example 1 and compute 1 period ahead forecast.