Update assignment_2.ipynb

[SainaAliabadi]

What changes are you trying to make? (e.g. Adding or removing code, refactoring existing code, adding reports)

Ran codes to find numerical variables and fit simple and multivariable linear regression to find the association of x variables with y.

What did you learn from the changes you have made?

Some variables have positive and negative associations with y, while there are some non-linear relationships of the x variable with y

Was there another approach you were thinking about making? If so, what approach(es) were you thinking of?

No

Were there any challenges? If so, what issue(s) did you face? How did you overcome it?

No

How were these changes tested?

I checked the results from models and plots with the actual dataset.

A reference to a related issue in your repository (if applicable)

NA

Checklist

• [ X] I can confirm that my changes are working as intended

[PatelVishakh]

Assignment 2: Complete, Great Work!
Suggested changes:

Q1) II) More relevant response is Mpg is a continuous response variable stored in our dataset as float64 type.

Q2) I) Also Should explain what Positive, Negative and No association imply about the relationship with the response. ie when model year increases on average the MPG goes down.

Q2) II) Should explain the plotted line represents the LINE OF BEST FIT with respect to the mean of sum of squared errors. These differences are called residuals, and they show how far each data point is from the predicted value on the line.

Q4) In the context of this module by RMSPE we are referring to Root Mean Square Prediction Error and not Percentage Error.