Business Forecasting regarding Seasonally Adjusted Sales - Statistics Project Example

The paper “Business Forecasting regarding Seasonally Adjusted Sales" is a worthy example of a statistics project on finance & accounting. Exponential smoothing is a moving average technique used in the analysis of time-series data as well in forecasting. We chose the best exponential models where recent collect data have more weight in making a prediction as compared to the older observation, we consider several factors like in this case where we have to consider whether the data is significantly influenced by a seasonal factor as well as trend. Secondly, we have to consider whether the data has a linear trend and lastly whether the time series problem is stationary.

In this case, the most suitable is the triple exponential smoothing (winters) due to the reason that sales are seasonal and also the sales in which the trend is being investigated are not stationary. In addition, to this in order to ascertain the smoothing contract, we will use the mean of the squared deviation of the forecasted values and the actual sales. In this case, we will calculate the mean of the squared sum of errors (Jensen & Bard, 2003).

In using the winter s method the following Alpha =0.3, Beta = 0.7 was used. These coefficients were obtained through the trial and error method with a minimum error of 0.04 and a maximum error of 1.95.

Forecasted sales for the period of jan2011 to December 2011,

d) Report
In order to ensure stability in the earnings of the firm, the management has a tendency of predicting the sales by use of times data available and this data is used in forecasting the future situation in the firm. In a motive to ensure that the sales of the organizations are maximized the firm decided to carry exponential smoothing of the sales forecasted bay applying the various methods like Double exponential (Browns), Double exponential ( Holts), single exponential smoothing, ARIMA model and the triple exponential smoothing.

 After careful analysis and consideration, the management decided to use the triple exponential method by winter. This was due to the reason the time series data concerned was seasonal in nature that sales vary with seasons. This method has a high level of accuracy, whereby it has to mean standard deviation is about 22.22 this showcase that this method is significantly feasible.

From the above discussion, an MSD 22.22 shows that the level of risk is minimum thus there method is sufficient in the prediction process. The winter’s method will predict sales of the twelve months with an accuracy of 88%, but through the prediction, the management is likely to face the following problems; validation of alpha value, the value was obtained from trial and error method this method is not significantly valid. Second, the prediction values are based on historical data which might not reflect the present situation in the market.

The forecast data will help in planning since the sales forecast will be used to calculate the market share that the company will control, thus if the company wishes to increase its market share the management will undertake an action that will lead to widening the market that firm sells its product. The forecasted sales will also help in determining the profitability of the firm in the future.

Although there were variations between the actual sales and predicted then these have been caused by normal sampling errors, random influences, and situations where future sales turn out to be radically different from the past.  Although this method is faced by the trail method in establishing the alpha and it also very complicated analysts expect the answers to be statistically feasible.

In conclusion, there is a need for forecasting to be used as a tool in planning since it gives a representation of future earnings of the firm thus it can be of help in plan formulation since it provides information that is almost true.