Week 6

Executive Summary

Starbuck's Inc business operations dataset (Net revenues from 2000 to 2018) is used in developing and analyzing the three quantitative forecasts. The three quantitative forecasts will be compared against each other to identify the best method for Starbuck Inc.

· Linear Regression Forecasting

· Simple Moving Average

· Exponential Moving Average

Linear Regression Forecasting

The linear regression technique is used to show a straight line to data in the past.

· Regression is a functional relationship between two or more correlated variables.

· Used for both time and relationship forecasting.

Y = B0 + B1X1, where B0 is the y-intercept, B1 is the gradient, X is the independent variable, and Y is the dependent variable (Box, Jenkins, Reinsel, & Ljung, 2015). The table below shows the line of best fit to the data set.

Figure 1: Graph of a linear regression forecasting

10 = X- bar

11358.47 = Y-bar

1208.01 = b

-721.67 = a

Regression Equation is Y = 1,208.01X + 721.66667

Simple Moving Average (SMA) quantitative forecast

The simple moving average quantitative method

· Used to represent an average that "moves" through the time series

· Arithmetic moving average obtained by calculating the average net revenues over a given number of periods.

· Computed from the company's closing sales.

· 3-year simple moving average will be computed by adding the annual net revenue for the last 3 years and dividing the summative by 3.

Table 1: Table of 3-year Simple Moving Average

The computation is repeated for each net revenue on the chart, and the average is calculated by dropping the oldest observation and adding the newest observation (Liu, & Li, 2015). The graph below is a plot which demonstrate data sequence in the table above. The SMA begins on year 2003 and continues. The number of periods in SMA is 3. Moreover, the values of mean squared error (MSE) and mean absolute deviation (MAD) and are 2561.46 and 8307494.37 respectively.

Figure 2: Graph of a simple moving average.

Exponential Moving Average (EMA) quantitative forecast

The Exponential Moving Averages is decreasing the lag by applying the weighting factor which reduces exponentially.

· The weighting factors is based on the most recent data as well as the total number of periods required.

· Three phases involved during the computation.

· First phase involves computing for SMA

· Second phase involve the computation of weighting factor

· Last phase involves the calculation of exponential weighted moving average

Table 2: Exponential Moving Average

The weighting factor applied to the most recent data depends on the number of period in the moving average. Weighting for a short period is preferred than for a longer period. Therefore, Starbuck`s 3-year period can be said to be more relevant. The chart below illustrates the sequence of data and the exponential smoothing from Table 2.

Figure 3: Graph of Exponential Moving Average

The best forecasts method for the company

Based on the three quantitative forecasts methods, exponential moving average (EMA) forecast technique is more effective and it should be employed in the company. Though this forecast method seems more complex when compared to the other two techniques, EMA forecast indicates a well-calculated forecast value which considers all factors of production within a company, hence producing an accurate and reasonable forecast. Moreover, the EMA technique lowers the lag which is available in SMA through advocating and implementing the use of shorter-predictive periods as a way of improving the company`s levels of accuracy.

Impact this forecast on the firm's financial metrics standpoint

The exponential moving-average forecast method will result in several positive impacts on Starbucks' financial metrics. The forecast technique will produce accurate and precise net revenue forecast which will aid the management sector to come up with more quality products and services catering to the needs of all the customers. Customers satisfactory is directly proportional to the company`s performance.

Furthermore, exponential moving-average forecast method will help Starbuck company in the creation of more effective adverts to attract the attention of more customers. Business companies should consider exponential moving-average technique as the best method for predicting the nature and performance of the company.

Conclusion

According to the three quantitative forecast methods, the exponential moving average appears to be the best method for Starbuck Company and other business companies as well. The EMA responds to the data fluctuation quicker than a simple moving average. Therefore, it produces accurate and price forecast.

References

Box, G. E., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). Time series analysis: forecasting and control. John Wiley & Sons.

Liu, D. J., & Li, L. (2015). Application study of comprehensive forecasting model based on entropy weighting method on trend of PM2. 5 concentration in Guangzhou, China. International journal of environmental research and public health, 12(6), 7085-7099..