Objectives of the paper and model specifications
The main objective of this paper is to carry out a regression analysis of consumer related data for a specific product. The product selected for analysis was sport utility vehicles (SUVs) sales in the United States. The United States Department of Transportation website was the source of the data used from the paper. It contained sales, market share and fuel consumption information. Using this relevant consumer information a linear regression model was developed that sought to investigate the relationship between SUV Sale, market share and fuel consumption. In this case, SUV Sale is the dependent variable and refers to the number of SUVs sold within the United States per year. On the other hand, the market share (refers to the total market share enjoyed by SUV vehicles in the United States market) and fuel consumption (average manufacturer specified fuel consumption for all SUV vehicles) are the independent variables.
Regression model:
The regression model developed was in the form:
Qx = a – bPx + cA + ..
In this case the regression model becomes:
Total SUV sales = a + b (market share) + c (fuel consumption)
Where,
a, b, c = regression coefficients (constants)
Total SUV sales = the number of SUVs sold within the United States per year
Market Share = the total market share enjoyed by SUV vehicles in the United States market (Number of SUVs sold / Total vehicles sold)
Fuel consumption = average manufacturer specified fuel consumption for all SUV vehicles sold in the United States
Description of :
As mentioned earlier, the data used for this paper was sourced from The United States Department of Transportation website. In order to locate the dataset a Google search was conducted with search terms that include ‘SUV’ ‘sales data’ ‘annual sales.’ Google returned a number of website sources, which were examined and sorted. From the list of searches, the best and most credible source was The United States Department of Transportation website. Using the search feature available on the website an additional search was carried out using similar search terms. The data titled Table 1-21: Period Sales, Market Shares, and Sales-Weighted Fuel Economies of New Domestic and Imported Light Trucks (Thousands of vehicles) was downloaded and It contained sales, market share and fuel consumption information for a number of light trucks (Bureau of Transportation Statistics, 2016).
Using Excel, the original dataset was filtered in order to remove the small, medium, and large size pickup data and only leave data on SUVs. The main reason for this is that the study was interested in studying market share and consumption as factors that affect sales of SUVs in America. The dataset contains annual data from 1990 to 2012. It also contains data from 1980 and 1985 (Bureau of Transportation Statistics, 2016). This in total consisted of 25 observations for each dataset. Using the filtered data set, the total market share for SUVs was calculated, as well as, the average consumption for SUVs. The data was then organized into a table and a regression analysis carried out in Excel. A correlation analysis was also carried out in order to test the association between the variables involved in the study.
The regression results also show that the prediction model is strong. This is mainly because the model summary indicates a high R-squared value equal to 0.8221. The R-squared value indicates that 82.21% of the dependent variable can be explained by the dependent variables. Examining the significance of the F value, it is evident that the regression model is statistically significant since the p value is less than alpha (0.05) (Sasvári, Agresti, Koop, and Clark, 2013, 2013, 2013, 2013).
The results of the regression model from the above analysis imply that market share and fuel consumption significantly influence SUV sales in the United States R2 = .8221, F (2, 24) = 50.83, p < .005.
Total SUV sales = a + b (market share) + c (fuel consumption)
a = 4438.124647
b = 87.28267621
c = -270.6105364
This implies that Total SUV sales can be predicted using the formula below:
Total SUV sales = 4438.12 + 87.28 (market share) - -270.61 (fuel consumption)
Examining the p values for the coefficients it is evident that the constant term and all predictors (market share and fuel consumption) are significant since they have a p value that it less than alpha (0.05) (Cleff, Quirk, Verschuuren, and Carlberg, 2014, 2015, 2014, 2014). The p value for the constant term is equal to 0.03, the p value for market share is equal to 0.000000088, and the p value for fuel consumption is 0.023. This implies that all the predictors are significant and are necessary in the predicting the total SUV sales (Vogt, 2012).
The nature of the relationship between the dependent and independent variables is evident from the regression results obtained (Afifi et al, Rencher et al, and Chatterjee et al, 2012, 2012, 2012). From the regression model, it is clear that the regression coefficient for market share is equal to 87.28267621. This value is positive and implies a positive relationship between market share and SUV sales. This implies that as market share increases so does SUV sales and vice versa. Additionally, the regression coefficient for fuel consumption is equal to -270.61. This value is negative and implies that there is a negative relationship between SUV sales and fuel consumption. This implies that as fuel consumption increases, SUV sales decrease.
Elasticities of demand:
Controlling fuel consumption (A)
(dQ / dP) = (P) / (a + b P + c A)
= (b P) / (a + b (P) + c (A)
a = 4438.124647
b = 87.28267621
c = -270.6105364
= 87.2 (market share) / (4438.1 + 87.28 (market share) – 270.61 (fuel consumption)
Therefore, for 2012
Market share = 58 %
Fuel consumption = 24.3
= 87.2 (58%) / (4438.1 + 87.28 (58%) – 270.61 (24.3)
= 50.75 / (4438.1 + 50.75 – 6575.8)
= 50.75 / -2086.95
= 0.0243
Conclusion:
In conclusion, the results obtained from the regression and correlation analysis are as expected. This is mainly because consumers consider factors such as fuel consumption and brand popularity before purchasing an SUV. Therefore, in turn this factors influence consumer demand and in turn affect sales. The negative relationship between total sales and fuel consumption shows that increases in fuel consumption are undesirable to customers. This is mainly because of the expense of driving the vehicle with high fuel consumption, as well as, the impact on the environment. Therefore, car manufacturers should ensure that their manufactured car models have low fuel consumption in order to attract more customers and increase sales.
The positive relationship between market share and fuel consumption is also as expected. In theory, higher market share implies increased brand and consumer reach, consumer loyalty and increase in demand, as well as, sales of a specific vehicle. This implies that SUVs with higher market share will sell more as compared to brand with limited consumer reach. Lastly, using the predicted regression formula, which is Total SUV sales = 4438.12 + 87.28 (market share) - -270.61 (fuel consumption) one can be able to predict future SUV sales using market share and fuel consumption.
References:
Afifi, A. A., May, S., Clark, V., & Afifi, A. A. (2012). Practical multivariate analysis. Boca Raton: Taylor & Francis.
Agresti, A. (2013). Categorical data analysis. Hoboken, NJ: Wiley.
Bureau of Transportation Statistics (2016) National Transportation Statistics. Retrieved 3/11/2015 from
Bureau of Transportation Statistics (2016) Search. Retrieved 3/11/2015 from http://www.rita.dot.gov/bts/search/google/SUV?query=SUV&cx=008370246888531060023%3Agwfrxpqlksk&cof=FORID%3A11&sitesearch&safe=medium
Bureau of Transportation Statistics (2016) Table 1-21: Period Sales, Market Shares, and Sales-Weighted Fuel Economies of New Domestic and Imported Light Trucks (Thousands of vehicles). Retrieved 3/11/2015 from http://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/publications/national_transportation_statistics/html/table_01_21.html
Carlberg, C. G. (2014). Statistical analysis: Microsoft Excel 2013. Indiana : Que
Chatterjee, S., & Hadi, A. S. (2012). Regression analysis by example. Hoboken, New Jersey: Wiley.
Clark, R. M. (2013). Intelligence analysis: A target-centric approach. Los Angeles: SAGE/CQ Press.
Cleff, T. (2014). Exploratory data analysis in business and economics: An introduction using SPSS, Stata, and Excel. Cham : Springer
http://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/publications/national_transportation_statistics/index.html
Koop, G. (2013). Analysis of economic data. Hoboken : Wiley
Lind, D. A., Marchal, W. G., & Wathen, S. A. (2015). Statistical techniques in business & economics. London: McGraw-Hill Education.
Quirk, T. J. (2015). Excel 2013 for business statistics: A guide to solving practical business problems. Cham : Springer
Rencher, A. C., & Christensen, W. F. (2012). Methods of multivariate analysis. New Jersey : Wiley
Richardson, R. (2015). Business applications of multiple regression. New York: Business Expert Press
Sasvári, Z. (2013). Multivariate characteristic and correlation functions. Boston : De Gruyter
Verschuuren, G. M. N. (2014). Excel 2013 for scientists. Uniontown, Ohio: Holy Macro! Books.
Vogt, W. P. (2012). Correlation and regression analysis. Los Angeles, Calif: SAGE.
Vogt, W. P. (2012). Correlation and regression analysis: 2. Los Angeles ˜[u.a.]œ: SAGE.
Vogt, W. P. (2012). Correlation and regression analysis: 3. Los Angeles ˜[u.a.]œ: SAGE.
Vogt, W. P. (2012). Correlation and regression analysis: 4. Los Angeles ˜[u.a.]œ: SAGE
Appendix:
