Skip to content

Performed statistical analysis using R on the MechaCar prototype production data to gain insights on the production issues for the manufacturing team

Notifications You must be signed in to change notification settings

teresa-le/MechaCar_Statistical_Analysis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

21 Commits
 
 
 
 
 
 

Repository files navigation

MechaCar Statistical Analysis

Linear Regression to Predict MPG

Vehicle_weight, spoiler_angle, and AWD's p-values are more than the significance level 0.05 so they likely contribute a random amount of variance to the mpg values in the dataset.

The slope of the linear model is not 0; the slope is the coefficient of the independent variable used to predict the dependent variable. For example, if vehicle_length is being used to predict mpg, then the slope would be 6.267. If vehicle_length increases by 1 and the values of the other independent variables are held constant, then mpg would increase by 6.267.

The linear model predicts mpg of MechaCar prototypes pretty effectively. The model can predict 71.49% of the observations in the model. Based on the p-value, we reject the null hypothesis; there is sufficient evidence that at least one of the independent variables significantly influences and can predict the dependent variable mpg.

Summary Statistics on Suspension Coils

Summary Statistics for All Lots

When it comes to all manufacturing lots, the suspension coils currently meet the design specification, which is to have a variance no more than 100 pounds per square inch.

Summary Statistics for All Lots by Lot

In the table above, the variance for lot 3 is 170.3 pounds per square inch so lot 3 does not meet the design specification.

T-Tests on Suspension Coils

All Lots

Lot 1

Lot 2

Lot 3

One sample t-tests were conducted to compare the sample means of PSI for all lots combined and each separate lot to the population mean of 1500 pounds per square inch.

The p-values for all lots combined, lot 1, and lot 2 is greater than the significance level 0.05. Therefore, we do not have sufficient evidence to reject the null hypothesis, and the means for the combined lots, lot 1 and lot 2 are statistically similar to the population mean.

The p-value for lot 3 is less than the significance level 0.05 so there is sufficient evidence to reject the null hypothesis and that the mean for lot 3's PSI is statistically different from the population mean.

Study Design: MechaCar vs Competition

In this study, we will be comparing MechaCar to similar cars belonging to AutosRUs’ competitors. I would like to see whether there is a statistical difference between the mean safety ratings that MechaCar receives in comparison to other cars.

Test: The metric that is going to be tested is the safety ratings of MechaCars vs. the latest models of similar vehicles from other manufacturers.

Null Hypothesis:The mean safety ratings for all of the cars are equal.

Alternative Hypothesis:At least one of the mean safety ratings is different from all of the other groups of cars.

Statistical Test: The statistical test that I am going to use to test the hypothesis is a one-way ANOVA; multiple samples are being compared (safety ratings of multiple different cars) and only one metric is being tested, which is the safety rating.

Data Requirements: The data that would be required to run the statistical tests would be samples of the safety rating scores that MechaCar and AutosRUs' competitor's cars received.

About

Performed statistical analysis using R on the MechaCar prototype production data to gain insights on the production issues for the manufacturing team

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages