Analysis Design and Assumptions

Analysis Design and Assumptions

In the evaluation of possible independent variables that may impact the dependant variable of vehicle sales, Nissan Vehicle Sales data for a representative Car (Sentra) and SUV (Xterra) was gathered for the cities of Nashville and New York by month for a 4 year period from 2003 from official Nissan Web-site and other sources.

Our Specific areas that were of interest were gas prices, population density, and average income. Gas Prices data was gathered using the “Petroleum Navigator” tool in the US Department of Energy (DOE) website and “gasbuddy.com”, which contains comprehensive listings of gas prices by different cities for different periods. Income and Population Data were from the 2000 US Census data results and specific Metropolitan City Web-sites.

Gas prices would be represented in the study by the prices of gas in the two cities specified over a four year period. The analysis would presume that as gas prices grow consumers would then purchase more fuel efficient cars of use some other form of transit i.e. public transportation. Thus, it would be presumed that there would be a negative relationship between vehicle sales and gas prices.

Population density would be measured as the population of Nashville and New York. As a result of the census being done every five years the population used was constant over our four year analysis of the data. In more heavily populated areas it is presumed that as the population increases the demand for fuel in that particular area would increase. Thus, it is also presumed that the greater the population density the higher the gas prices would be in a city.

As median incomes become higher, it is presumed that populations would not change their buying preference based on the price of gas. As such, there is expected to be a positive correlation between per capita income and vehicle sales.

Analysis of Results

During the initial analysis, the values were input into scatter plots with the dependant variable on the Y-axis on all graphs. Also, a linear regression line of estimation is depicted in each graph. Upon viewing these graphs, the population variable was excluded from the analysis because it was a constant.

Summary of Findings

Effect of Gas Prices on Vehicle Sales by Type (Car vs. SUV)

Vehicle Type Nashville (Beta) New York City (Beta)

Car (Sentra) +.199 +.240

SUV (Xterra) -.641 -.672

Nashville:

The Beta for Car (Sentra) is +.199 indicates that when Gas Prices (independent variable) go up by 1 Standard Deviation then Car Sales (dependent variable) goes up by .199. Similarly, the Beta for SUV (Xterra) of -.641 indicates that when Gas Prices (independent variable) go up by 1 Standard Deviation then SUV Sales (dependent variable) goes down by .641.

Hence, upon analyzing the aforementioned observations, it can be deduced that when Gas Prices go up then Car Sales go up while SUV Sales go down. This is line with our hypothesis that increasing Gas Prices affect (negatively) the Sales Volume of SUVs while increasing the Sales Volume of more fuel efficient cars.

New York City:

The Beta for Car (Sentra) is +.240 indicates that when Gas Prices (independent variable) go up by 1 Standard Deviation then Car Sales (dependent variable) goes up by .240. Similarly, the Beta for SUV (Xterra) of -.672 indicates that when Gas Prices (independent variable) go up by 1 Standard Deviation then SUV Sales (dependent variable) goes down by .672.

Hence, upon analyzing the aforementioned observations, it can be deduced that when Gas Prices go up then Car Sales go up while SUV Sales go down. This is line with our hypothesis that increasing Gas Prices affect (negatively) the Sales Volume of SUVs while increasing the Sales Volume of more fuel efficient cars.

Conclusion:

It can be observed here that higher Gas Prices affect the Sales of SUVs negatively