Business Statistics for Financial Decision †MyAssignmenthelp.com

Question: Discuss about the Business Statistics for Financial Decision. Answer: Introduction: The report analyses sales figures of Schmeckt Gut for the last 25years (1991-2015)and tries to forecast sales for the year2016. The first explanatory variable considers here is the Gross Domestic Product in US dollar capture the income development. The trend in prices is indicated in terms of average increase in the price index. Population is another important factor determined the sales of energy bar. The data on population for the age limit 15 to 65 years are studied. A survey is conducted to measure the level of satisfaction from the energy bar consumption. The satisfaction level is ranked from 0 to 10 where 0 implies not satisfied and 10 implies very satisfied. In order to promote products company makes advertisement. Advertisement of energy bars that is number of advertisement on an average. The last variable consideris the number of stores from where energy bars can be purchased. All the chosen explanatory variables are likely to have large influence on sales. Sales US$ Survey score Advertisement Stores Sales US$ 1 Survey score 0.588601019 1 Advertisement 0.986236917 0.54833082 1 Stores 0.971109284 0.52890297 0.970966786 1 The correlation matrix shows the correlation between sales and independent variables of satisfaction score, number of advertisement and number of stores. The correlation coefficient between sales and satisfaction score is 0.59. The positive correlation implies a positive relation between sales and survey score. The coefficient is high showing a strong correlation between the two variables. The correlation coefficient between Sales and number of advertisement is 0.99. A value of correlation coefficient close to 1 shows a perfect positive relation between the variables. The correlation matrix gives value of correlation coefficientequals to 0.9711. This value is also close to 1 implying perfect linear relationship between sales and stores. Time Series Regression Regression Statistics Multiple R 0.997898471 R Square 0.995801359 Adjusted R Square 0.994401812 Standard Error 0.020094198 Observations 25 ANOVA df SS MS F Significance F Regression 6 1.7238 0.2873 711.5169 0.0000 Residual 18 0.0073 0.0004 Total 24 1.7310 Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Intercept 12.763 1.008 12.660 0.000 10.645 14.881 ln(GDP) 0.065 0.052 1.267 0.221 -0.043 0.173 ln(Price index) -0.131 0.017 -7.615 0.000 -0.167 -0.095 ln(Popualtion) -0.350 0.135 -2.586 0.019 -0.635 -0.066 ln(satisfaction) 0.084 0.025 3.413 0.003 0.032 0.136 ln (advertisement) 0.868 0.093 9.314 0.000 0.672 1.064 ln(Stores) 0.230 0.078 2.954 0.008 0.066 0.393 From the regression result the estimated equation is obtained as The coefficient of ln(GDP) is 0.065.The positive value of the coefficient implies a positive relation between sales and GDP. The variable is not statistically significant as the p value is 0.221, which is greater than the significance level of 5%. The coefficient of ln(Price Index) is -0.131. This indicates 1% increases in prices reduces sales by 0.13%. The variable is statistically significant as indicated by the significant p value of 0.000. The coefficient of ln(population) in -0.350. The unit increases in population aged between 15 to 65 years causes a decrease in sales by 0.35%. The variable ln(population) is statistically significant. The three remaining variables satisfaction score, number of advertisement and number of stores are significant determinant for sales of energy bars. These three variables have positive impact on sales. However, the highest effect is estimated for number of advertisement with coefficient value of 0.868, followed by number of stores and satisfaction score with estimated coefficient of 0.23 and 0.084. The following forecasts are given for the independent variables. Based on these information sales in 2016 can be predicted. If GDP grow by 2.5%, then GDP becomes Price index = 2%. Given population grows by 0.5 percent population in 2016 is Satisfaction score = 7.5 Number of advertisement = 18 Number of stores = 12 Taking the logarithm of each of the dependent variables and putting them in the estimated equation the sales of 2016 is predicted as Taking antilog, the predicted sales in 2016 is obtained as 1047387.7 Therefore, the approximate sale of Schmeckt Guts energy bars in 2016 is 1047387.7. Methods of forecasting The alternative forecasting techniques that can be applied are trend analysis and exponential smoothing. Trend analysis is a common forecasting technique used by business or other organization to predict the future outcome based on historical data. In statistics trend analysis captures the pattern of time series behavior. Regression analysis gives a cause and effect relation based on least square measures (Cameron Trivedi, 2013). Trend analysis can predict the future value without the estimated equation. It analysis the behavior of variables overtime and then predict the future value. In this study trend of sales and the dependent variables from 1991 to 2015 and the forecasted value of these indicators are used to predict sales in 2016. Accordingly the predicted sale of 2016 is obtained as 1050012.9. The predicted value of sales by trend analysis is very close to that obtained from the regression analysis. Exponential smoothing is a kind of moving average used for time series forecasting. The forecasting is done using the following equation Ft is the forecasted sales of year t At-1 is the actual sales of previous year Ft-1 is the forecasted sales of the previous year is the smoothening constant , 01 The forecasting is done for a given value of . As no value of is given, it is taken as 0.5. This forecasting technique compares the prior forecasting estimate with actual value and use the difference or error to make new forecast (Montgomery, Jennings Kulahci, 2015). Here values of baseline variable are used as a medium of forecasting. In the exponential smoothing previous years sales value are used to forecast sales in 2016. The forecasted value of sales in 2016 is 898035.5. References Cameron, A. C., Trivedi, P. K. (2013).Regression analysis of count data(Vol. 53). Cambridge university press. Montgomery, D. C., Jennings, C. L., Kulahci, M. (2015).Introduction to time series analysis and forecasting. John Wiley Sons.