Forecasting involves the process of making predictions of the uncertain future with the emphasis being on the past as well as the present data through the use of trends. A typical example is the forecasting of future demand. Prediction is one of the essential techniques used in demand planning as well as budgeting analysis. With the demand or quantities produced in the previous periods, it will be possible to determine the forecasts. There exist two types of projections which is qualitative forecasts and quantitative forecasts. Some of the examples of qualitative forecasts include the Delphi Technique and the nominal group technique. However, this analysis will narrow down to the quantitative forecasting techniques. The three quantitative approaches include the trend, weighted moving average, and the exponential forecasting approaches.

Trend Forecasting

The trend is one of the conventional methods of forecasting, and it is applicable under the operations forecasting. Based on the existing patterns, the demand for future periods could be predetermined. It is quantitative forecasting as it is based on tangible and concrete data from the past. This forecasting technique uses the time series data with numerical values that are known during different points in time. The statistical data is plotted on the graph with time plotted on the x-axis, and the information being predicted graphed on the y-axis for instance sales. The data yielded from time series will be necessary as they will guide the forecaster in coming up with the forecasts for the future periods (Krige, 2016). A trend could be determined based on a particular constant pattern that has been reported on an organization with the variation being negligible.

Weighted Moving Average

Overall, this method applies a predetermined average weight to each month of the past data, summing up the data from each of the periods and dividing by the total sum of the weights. If the forecaster adjusts the weights such that their sum is 1, then the weights will be multiplied by the actual demand of each of the period applicable. These results are then summed up to achieve a weighted forecast (Lawrence et al., 2009). The more recent the data, the more the weight assigned to such forecasting data with the contrary being true. Using the demand example, a weighted average series using weights .4, .3, .2 & .1 are likely to yield forecasts for July as 45(.1) + 58(.2) + 62(.3) + 46(.4).

Exponential Smoothing Technique

A complex form of the weighted moving average method of forecasting is the use of exponential smoothing. It is named so as the weights often fall exponentially with the data aging. This method takes the forecasts from the previous period and adjusts it by a given predetermined smoothing constant (alpha) which is a value less than 1 – this is multiplied by the given difference in the previous forecast and the demand that is reported during the last forecast period. It is usually referred to as the forecast error (Pérez, 2017). Exponential smoothing extension can be used in an instance where the time-series has exhibited a trend that is linear .It isknowno as double-smoothing or Holt’s Model.

Comparing and Contrasting

Variation is likely to be experienced when using these forecasting technique sbecaust the data that is involvedis expectedyin changing substantiallyyIts is an issue that affects all the three forecasting techniques as the data used is usually data that is collected from the estimates. However, the trend analysis couldbe entirelye different as it will involve the use of time series analysis which at times always factor in the aspect of variation in the data duringdifferent periodse.

Exponential smoothing is also another technique often used in predictions. However, it is generally believed to be more accurate as compared to the weightedaverage methode as it involves theuse of  an alpha value which is the forecasting error. Therefore, this generally means that this is one of th epractica emethod,s which will be used in the accurate presentation of forecasts.

All the approaches use data from demand orother relevantt market data which they effectivelymanipulate too come up withthe usefule trends as well as analysis. Additionally, all these techniques require some element of mathematical manipulation so that the relationship between the data can be determined.

Out of the three techniques, the most effectivemethode is the exponential smoothing forecasting technique. A financial institutionwouldl, for instance,e want to determine the amount of loans that they are likely to lend their customers in the second quarter ofthe fiscall year. They will need data relating to the previous loans givenout inn thelast quarterss and assign them weight based on the recent data withthe latestt data being given muchpressuret and the contrary applying to the old data.

References

Krige, D. G. (2016). Two-dimensional weighted moving average trend surfaces for ore evaluation. S.l.: South African Institute of Mining and Metallurgy.

Lawrence, K. D., Klimberg, R. K., &Lawrence, S. M. (2009). Fundamentals of forecasting using Excel. New York, N.Y: Industrial Press.

Pérez, A. C. R. A. (2017). A double exponentially weighted moving average control chart for the individuals based on a linear prediction.

 
Do you need high quality Custom Essay Writing Services?