Demand Estimation Bread Company
Dr. Jen-Hsiang Lin
Demand Regression Model Estimation
The preferred model is:
Q = b0+ b1P +b2P +b3Ad +b4I
Table 1: Regression model
The independent variables are all statistically significant in this model. The dependent variable, in this case being demand, against the independent variables are: price, competitor price, advertising and income of the households surveyed. When applying the regression of demand against the other variables we get the following:
Q = 128832.2 -19875.95 Price + 15467.94 Competitor Price + 0.261Advertising +8.78 Income
The San Francisco demand equation reveals that the demand, as independent variable reveals has a inverse relationship between price and demand. The coefficient of price has a negative value (-19875.95). Therefore when the average price per meal raises the number of meals served declines.
On the other hand, the competitor price as a positive effect to the San Francisco demand equation. In effect there is a direct relationship between competitor’s price and demand. The coefficient of price has a positive value (15467.94). factor of 15467.94. Thus when the average price charged by competitors increases by 15467.94, the number of meals served escalates one unit.
The advertising component does not have a significant effect on the demand as shown in demand equation given that only less than one-unit increase in the number of meals served in the household (0.261) per every unit increase of advertisement budget.
The average income per household in the outlet service area covered by the company reflects an increase in the number of meals served by a factor of 8.78 when the average income increases by a unit.
The coefficient of determination (R2) value gives us the percentage change explained by independent variables that have an effect in the dependent variable. In this scenario the coefficient of determination is 0.833. Therefore 83.3% of the variations we see in the dependent variable are explained by the independent variables. We can confidently assume that the model works well.
Question B :
San Francisco Bread Company should consider targeting high-income customers because per every unit increase in income there is an increase in demand by 8.78. Nonetheless, before taking the decision, we must take into consideration the income elasticity of demand for bread to determine the proportional change
income elasticity of demand= ?Q?I*IQ =8.78* 51044598412 = 0.74This tells us that an increase in income leads to a less than proportionate increase in quantity of bread consumer and thus, focusing on higher income consumers might not increase the revenues of the company. Furthermore it is important to highlight the fact that according to the descriptive statistics, income levels have a low range, with the minimum being 46017 and the maximum 5545, meaning that income disparity is insignificant.
Question C: cross elasticity of demand= ?Q?Px*PxQ =15467.93* 6.16598412 = 0.159The prices of the competitor do not have much of a proportionate impact on the quantity of sales. So discounting by the competitor won’t have a significant impact on San Francisco Company’s sales.
Based on this, San Francisco Company should not be concern about the discounts imposed by competitors but it is still important to keep track of any price movements they make.
elasticity of demand due to adverstisement = ?Q?Ad*PxQ
=0.2607* 244649598412 = 0.106Clearly the advertising strategies by San Francisco Company are ineffective because they are not giving significant results. The regression model tells us that the advertising has a very small impact on the quantity of meals consumed. Additionally, it has a very small proportionate increase in the number of meals served in the household (0.106) as product of a one unit increase of the advertisement budget.
elasticity of demand= ?Q?P*PQ = 19875.95 * 6.93598412 = -0.23San Francisco Bread Company should absolutely consider offering discounts to the customers. This action could potentially increase their market share by taking advantage over its competitors. Discounting the prices of the product will result to some consumers shift from utilizing the competitor’s products to theirs.
Log linear demand equation
LQ=?+?1P+ ?2Px+ ?3Ad+ ?4I+ ?Hence:
LogQ=12.5114-0.0332P+ 0.025Px+ 0.000000432Ad+ 0.0000148IThe intercept is very significant given that its P value is less than 0.05 and represents the expected mean for the log Q. b1 = 0.0032 and which means that every unit increase in price results in 0.32% (since the exp (0.0032) = 1.003205), fall in quantity demanded. This is higher the elasticity calculated beforehand. b2 = 0.025669 implying changes in competitor’s prices result to 2.6% change in quantity of san Francisco’s bread consumed, lower than the 0.159 calculated earlier. The exponential variables calculated for advertising and income are negligible very small. In conclusion there is a significant change in the elasticities in relation to the ones on demand equation