Trendless data can be evaluated with the help of single exponential smoothing technique. This technique can be implemented on time series data to make the forecast or to smooth the data for presentation. It exponentially decreases the weight over the time and can be implemented for analyzing the financial market or economic data (Guire 2011).
Centered moving average can be computed by averaging the past and future data for a given time point but at the time of forecasting the future is typically unknown therefore this method cannot be used for forecasting. On the other hand, trailing moving average can be set on most recently available data set hence it is best suitable for the forecasting of the information (Shmueli and Bruce 2010).
Exponential smoothing is used for analyzing the data based on the time series to forecast the future trend of information. The exponential smoothing can be solved using formula, in which the value of α ranges from 0 to 1, depending upon its damping rate (Hyndman2008).
In the exponential smoothing, the value of α is chosen depending on the degradation in the value of the given data. For little smoothing the value of the constant α will be less, i.e. can be range from .01 to .15, on the other hand, for moderate smoothing the value of α can range from .40 to .50. As small value of α give little weight to the most recent time period while provides greater weight to the past unit time period. This is because the value of weight is decreased by (1- α), thus small value of α effect less (Guire 2011).
If α=.05, then Ft+1=.05yt+.95ft (heavy smoothing, low adaption)
If α=.20, then Ft+1=.20yt+.80ft (moderate smoothing, moderate adaption)
If α=.50, then Ft+1=.50yt+.50ft (little smoothing, quick adaption) (Doane 2006).
Doane, (2006) Applied Statistics In Business And Economics. USA: McGraw-Hill Professionals.
Guire, G. M. (2011) Handbook of Humanitarian Health Care Logistics. Austria: George Mc Guire.
Halder, C. (2010) Power System Analysis: Operation And Control. South Africa: Pearson Education.
Hyndman, R. J. (2008) Forecasting with exponential smoothing: the state space approach. Germany: Springer.
Kurtz, D. L. (2010) Contemporary Marketing 2011. UK: Cengage Learning.
Shmueli, G. and Bruce, P. C. (2010) Data Mining for Business Intelligence: Concepts, Techniques, and Applications in Microsoft Office Excel with XLMiner. USA: John Wiley and Sons.
Blog Post Provided By - Assignmenthelpexperts.com
Contact Us Email Id - Info@assignmenthelpexperts.com