A Sales Forecasting

Analysis of previous information is necessary for forecasting changes in market performance indicators (Box, Jenkins, Reinsel & Ljung, 2015). Analysts must create models that make use of the statistical data gleaned from market research in order to produce accurate predictions of market performance. Regression models can be used by WidgeCorp Company to forecast changes in monthly sales. Since numerous factors other than price affect a product's sales volume, WidgeCorp's monthly sales can be forecast using either a simple or multiple linear regression model (Guo, Wong & Li, 2013). However, to produce accurate results and avoid cases of variable omission bias in the model, a multiple linear regression model will be more applicable. This is because, a multiple linear regression model helps in forecasting an independent variable whose value is influenced by more than one explanatory variables.

When using simple linear to forecast monthly sales, the model will only have two variables that is sales volume and the prices of the beverages. An example of the simple linear regression model will be St= α - βPt, where St is the sales volume at month t, Pt is the price of cold beverages at month t, α is the autonomous sales (sales at zero price), and β is the marginal propensity to consume. However, this model does not include all the factors that influence monthly sales. Therefore, a multiple linear regression model in the form of Yt=a+bX1+cX2+…+nXn+ ε. Yt is the value of dependent variable at time t, a, b,..n are the coefficients , X1..Xn are the explanatory variables, and ε is the error term.

In order to come up with a regression model that does not suffer from variable omission bias, the following variables will be taken into consideration.

Monthly Sales (St)

The volume of sales at every month t will be the response or dependent variable in the model. This is because changes in other variables will affect how it changes.

Price (Pt)

The price charged by WidgeCorp on their beverages will dictated the volume of sales. An increase in the price will reduce sales while and decrease in prices will increase sales. Price will have a negative coefficient in the model.

Price of Related Beverages/Substitutes (PRt)

Since consumers of the cold beverages will consider shifting to substitutes whenever WidgeCorp increase their prices, the price of related products (substitutes) will be act as explanatory variables in the model. An increase in the price of substitutes will increase the sales volume while a decrease in price for substitutes will reduce the monthly sales volume (Guo, Wong & Li, 2013). Hence, the variable has a positive coefficient.

Marketing Cost (Mt)

Given that WidgeCorp is venturing into new products that they have not sold before, marketing cost directed to advertisements and product campaigns will influence the total monthly sales (Davis, Lockwood, Pantelidis & Alcott, 2013). Thus, monthly marketing cost for the cold beverages will influence the monthly sales volume thus; will be included in the regression model as an explanatory variable. Marketing budget is directly related to sales volume, hence a positive coefficient

Number of Distributors (Dt)

Guo, Wong & Li (2013) argues that WidgeCorp needs to reach the entire market through distribution of their cold beverages. The higher the number of distributors, the higher the sales volume of the beverages. A positive relationship exists between monthly sales volume and the number of distributors.

Based on the above variables, the multiple regression model for forecasting monthly sales will take the form of;

St= α0 - α1 Pt+ α2 PRt+ α3 Mt + α4 Dt + ε.

Where St- Sales at month t

Pt- Price of cold beverages at month t

PRt- Price of substitutes at month t

Mt- Number of distributors

α0, α1, α2, α3, and α4 are the regression coefficients

ε- Error term













References

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

Davis, B., Lockwood, A., Pantelidis, I., & Alcott, P. (2013). Food and beverage management. Routledge.

Guo, Z. X., Wong, W. K., & Li, M. (2013). A multivariate intelligent decision-making model for retail sales forecasting. Decision Support Systems, 55(1), 247-255.