# Linear regression: Parameters exercise

The graph below plots 20 examples from a fuel-efficiency dataset, with the
feature (car heaviness in thousands of pounds) plotted on the x-axis and the
label (miles per gallon) plotted on the y-axis.

**Your task:** Adjust the **Weight** and **Bias** sliders above the graph to
find the linear model that minimizes MSE loss on the data.

**Questions to consider:**

- What is the lowest MSE you can achieve?
- What weight and bias values produced this loss?

#### Click the plus icon to see the solution

The optimal linear model for this data has an MSE of 3.37, with a
weight of --0.12 and a bias of 16.96, as shown in the following image.
![Plot of 20 points and the optimal linear regression line for
these points, using the weight and bias values above.](https://developers.google.com/static/machine-learning/crash-course/linear-regression/images/parameters-exercise-solution.png)