# 56467

The general discussion regarding time series decomposition. What is time series decomposition? Why would a forecaster choose this method versus some of the other methods we've learned in class? ****In your answer, please make sure to define the components of a time series. Also, please feel free to elaborate upon time series composition topics that you found most interesting. Forecasting Methods Class I provide an example: Please don't copy this example. Thanks!What is time series decomposition? Time-series Decomposition is a model that is used by forecasters to isolate the existing time patterns within a data set, such as long-term or secular trends, seasonal trends, cyclical trends and irregular/random trends. The model breaks these series into its different parts and reassembles them to construct a forecast. (p. 298) The objective of this model is to decompose a series into individual components. The first step with this model is to remove the short-term variations (seasonal trends and irregular trends) from the data by applying an appropriate moving average model to the series. This component is described as deseasonalizing. Once the long-term component has been identified by applying a simple linear equation then you must measure the cyclical component with a cycle factor. This cycle factor is the most difficult component but if done right this factor can offer a clear understanding of what the data series is trying to tell us and can add to the accuracy of our forecasting. Why would a forecaster choose this method versus some of the other methods we've learned in class? This particular model would benefit a forecaster when the data series is composed of a large amount of historical data that encompasses decades of information. This model would be good to use when trying to forecast sales for a business in any industry, or could possibly be used to forecast a nations population. I think all the components of this time series decomposition was interesting. I have a feeling we will be using these components on our group assignment. One thing that I did find intriguing were the components of the composite indices for a business cycle. The different components introduced on (p. 311) are all very interesting factors that can be easily overlooked if you do not know what you are looking for.