AutosRU is developing a new prototype vehicle, the MechaCar. Data has been collected on various factors which impact the mpg and the suspension. This analysis performs a statistical analysis against these factors to determine how they impact the MechaCar.

Input: MechaCar_mpg.csv, Suspension_Coll.csv
Software: R, RStudio
Output: GitHub ReadMe

The following are the results of the statistical analysis.

A linear regression analysis was run against the MPG data. Using a significance level of 0.05 the results show that vehicle weight, spoiler angle and AWD likely have little to no statistical impact on MPG, while vehicle length and ground clearance do have a statistical impact on MPG. The overall p-value for the linear regression is very small (5.3x10^-11) which indicates that the slope of the regression line is not zero (indicating a correlation between the factors and MPG). In addition, the r-squared value is high (0.71) which indicates a strong overall correlation between the data and the MPG. The full results of the linear regression are below for reference.

An analysis was run against the suspension data to ensure the variance of the suspension coils not exceed 100 pounds per square inch per the design specifications for the MechaCar suspension coils. When looking at all manufacturing lots together the variance is less than 100 PSI and therefore within design specifications. However, when looking at the 3 manufacturing lots individually Lot 1 & 2 have very small variances which are well within the design specifications (variance = 1.0 and 7.5 respectively), but Lot 3 exceeds the variance and needs to be examined in more detail (variance = 170.3). The full summary statistics for suspension PSI are below for reference.

Further analysis was run against each of the lots to see if they were statistically different than the overall population.

• Lot 1 & 2: The p-value for lots 1 and 2 (1.6x10^-11 and 5.9x10^-4 respectively) were lower than the significance level and therefore are statistically similar to the overall population.
• Lot 3: The p-value for lot 3 (0.16) was much higher than the significance level and therefore IS statistically different than the overall population. The full t-test results are below for reference.

Additional analysis is recommended to compare the MechaCar to the competition to ensure it will be competitive in the marketplace. Since cost is a primary driver for most consumers and one of the easiest metrics for consumers to compare the competition, I am recommending further analysis on this metric.

• Metric to be analyzed: Cost
• Null Hypothesis: The mean cost of all the competition vehicles is statistically similar
• Statistical Test to be Used: One-Way ANOVA test, since we are testing one variable (cost) across many groups (competition vehicles)
• Data Needed for Test: To run the analysis we will first need to determine what vehicles are our competition. Then we will need a sampling of the cost for all of those vehicles across the population. Ideally the samples should be randomly generated across different location and dealerships.