Data source: MechaCar_mpg.csv

R Script: MechaCarChallenge.R (Deliverable 1)

1. Which variables/coefficients provided a non-random amount of variance to the mpg values in the dataset?

   The amount of non-random variance (i.e. high correlation) can be determined from the Estimate Coefficient of each variable. With this in mind, it is evident that both the vehicle_length and ground_clearance present high degrees of correlation with the mpg. This is indicated by their large and positive Estimate Coefficients.

2. Is the slope of the linear model considered to be zero? Why or why not?

   Given that the p-value = 5.35e-11 << 0.05, the model suggests that the null hypothesis should be rejected. Therefore, the slope of the linear model should NOT be considered zero.

3. Does this linear model predict mpg of MechaCar prototypes effectively? Why or why not?

   This model is somewhat effective at predicting mpg outcomes given these variables. Considering the Multiple R-squared: 0.7149 value, it is estimated that the model will accurately predict ~71.49% of the mpg outcomes./>

Data source: Suspension_Coil.csv

R Script: MechaCarChallenge.R (Deliverable 2)

The design specifications for the MechaCar suspension coils dictate that the variance of the suspension coils must not exceed 100 pounds per square inch. Does the current manufacturing data meet this design specification for all manufacturing lots in total and each lot individually? Why or why not?

Based on the total_summary table statistics, it would seem that the variance complies with this requirement (i.e. the variance of the suspension coils is 62.29356 < 100). However, upon closer inspection of each lot in lot_summary table statistics, it is clear that really only Lot1 and Lot2 meet this design requirement. Lot3 is throwing the total statistical interpretation off by its variance value of 170.2861224 (which is >> 100).

Data source: Suspension_Coil.csv

R Script: MechaCarChallenge.R (Deliverable 3)

All Lots:

Lot1:

Lot2:

Lot3:

Results Summary:

• Lot1 presents a highly reliable statistical match to the presumed mean PSI values across the population. The t-test value of 0 suggests that there is no statistically significant difference between the mean PSI of the sample (i.e. Lot1) distribution and the population distribution. A p-value of 1 indicated that the null hypothesis should be accepted (because it is >> 0.05).
• Lot2 also (but not as drastically) presents a reliable statistical match to the presumed mean PSI values across the population. The t-test value of 0.51745 suggests that there is little statistical difference between the mean PSI of the sample (i.e. Lot2) distribution and the population distribution. A p-value of 0.6072 indicated that the null hypothesis should be accepted (because it is > 0.05).
• Lot3 presents a lowly reliable statistical match to the presumed PSI values across the population. The t-test value of -2.0916 suggests that there is a significant statistical difference between the mean PSI of the sample (i.e. Lot3) distribution and the population distribution. A p-value of 0.04168 indicated that the null hypothesis should be rejected (because it is < 0.05).
• Looking at the T-test results across all lots, there can be observed a moderately reliable statistical match to the presumed PSI values across the population. The t-test value of -1.8931 suggests that there is quite a significant statistical difference between the mean PSI of the sample distribution and the population distribution. This is evidently being heavily distorted by the Lot3 mean. Still, the p-value of 0.06028 indicated that the null hypothesis should be accepted (because it is > 0.05).

(Deliverable 4)

Write a short description of a statistical study that can quantify how the MechaCar performs against the competition. In your study design, think critically about what metrics would be of interest to a consumer: for a few examples, cost, city or highway fuel efficiency, horse power, maintenance cost, or safety rating.

• What metric or metrics are you going to test?

  The following metrics could be supplemented to provide further investigation of MechaCar’s standing on the automotive market:

  1. Manufacturer

  2. Engine/Transmission Type: e.g. gas/automatic, gas/manual, hybrid, or electric.

  3. USD-per-Mile: to assess cost efficiency across both electric and gas vehicles.

     (Alternatively, could incorporate kWh/mile as an electric vehicle metric against mgp.)

  4. Annual repair/maintance costs

  5. Market price

  6. Mileage reached at end-of-life

• What is the null hypothesis or alternative hypothesis?

  Null hypothesis:

  MechaCar is a worthwhile investment due to the savings MechaCar owners achieve over the car’s life compared to other vehicles on the market.

• What statistical test would you use to test the hypothesis? And why?

  I would perform a multiple linear regression across these variables to find correlations and isolate variables that are significantly contributing to “market performance” outcomes. To justify these correlations, I would also perform a series of T-tests of each variable across manufacturer groups (and perhaps also across market price subgroupings), thus substantiating any outlying statistical influences.

• What data is needed to run the statistical test?

  We would need a large dataset including many relevant/similar manufacturers and consumer data. This would need to be obtained either through consumer reports or manufacturer data collections on their vehicle sales and statistics.