Sales Prediction for Northern Household Goods - Case Study Example

Sales Prediction for Northern Household Goods Sales Prediction for Northern Household Goods The linear regression models are used in order to determine the link between the two variables. The same technique can be applied to draw a scatter diagram as well. The sales of an organization are known to be significantly dependent of income levels of its target audience (Watson, 1964). The linear equation takes the form of a first degree mathematical statement which has a constant that is represented with a and b represents the level of change which is taking place in the predicted variable due a change of one unit in the predictor.

The linear regression is the commonly used model in management sciences while its application is significantly noted in the areas of social sciences and natural sciences as well. However, in simpler terms, the statistical technique is used in order to determine the level of predicted variable on the basis of each amount of the predictor. Furthermore, the application of bipolar regression is dying out because one outcome variable in indeed influenced by multitude of the inputs such as the level of one’s organizational commitment is influenced by supervisory behaviors of the company, his or her pay-scale and nature of the job.

In the given case, an organization is interested in developing a scatter diagram of the relationship between retail sales and disposal incomes of the family then the company is also looking to develop a linear regression model between the abovementioned variables. The 95% confidence interval about population data’s slope and finally, it is required to determine the level of sales if income of the household remains constant at the level of $58000. Scatter Diagram Linear Regression Equation Following is the linear regression table that is generated from applying SPSS on the given data set.

CoefficientsaModelUnstandardized CoefficientsStandardized CoefficientstSig.95.0% Confidence Interval for BBStd. ErrorBetaLower BoundUpper Bound1(Constant)1922.393274.9496.992.0001348.8582495.927X Income.382.025.95915.084.000.329.434a. Dependent Variable: Y Retail SalesWith the help of using the value of un-standardized coefficients we can build the following equation of linear regression: -Y = 1922.39 + 0.38 (X) -------------------------------------------- (i)The p-value associated with X as a predictor of Y highlights that the relationship between the two variables is statistically significant.

The lower and upper bound values determined in the preceding table for B represent a following 95% confidence interval: -0.329 < = B < = 0.434 Finally, the value of sales if the household income stands at the value of $58000 can be determined by putting the figure in (i) in the place of X. Y = 1922.39 + 0.38 (58000) = 23963 units. In the light of established linear model, we can assume that the company can sell 23963 units if the household income would be 58000. However, we can establish that sales can be successfully predicted with the help of entering the level of household incomes of the audience.

References Watson, G. S. 1964. Smooth Regression Analysis. The Indian Journal of Statistics Vol 26 No.4, 359-372.