Hands-on option

Basic stage for the hands-on option

Read about multivariate linear regression and play with this online tool to see how it works for two to four input variables. Note that these are assumed to be quantities, so you could experiment with the number of bedrooms, the number of bathrooms, and the living area in square feet, for example.

In-depth stage for the hands-on option

Experimenting either with the in-class example or a computational tool of your choice (Python and JavaScript are both fine, and R is another fine choice for this), look into how to fix a situation in which the scatter plot of the horizontal variable versus the vertical one does not form a straight line but instead a well-behaved curve of some sort (such as plotting the height of a person against their weight, which is a non-linear relationship).

Document your experiments and findings in writing with screenshots, including any code you might have used and all references.

Hint: the trick is called a transformation and it needs to be a mathematical function that has a known inverse (for taking a square root, the inverse would be taking the square, whereas for taking a logarithm, the inverse would be exponentiation). You can read more about this if you are interested in how the math behind it works.

Conceptual option

Basic stage for the conceptual option

Experiment with how adding more rows of data affects the way the model fits the data and the recommendation it gives to a value consulted. Read about linear regression and interpolation and reflect upon how what the class example is doing in terms of those two concepts.

What happens if you consult values that are outside of the range of the observations? Look into the concept of extrapolation and discuss.

In-depth stage for the conceptual option

Read about multivariate regression and identify how it differs from the simple linear regression we worked with in the class example.

Look into how to include qualitative information such as the neighborhood or the type of lodging (house, condo, etc.) into a multivariate linear regression model. Summarize your findings in your own words. Remember to clearly cite all sources.