How Far Off Is the Delivery Model, Typically?

A dispatcher uses a fitted regression model to predict how many minutes each food delivery will take. For five completed deliveries she records the time the model predicted and the time the delivery actually took:

Delivery	Actual $y$ (min)	Predicted $\hat{y}$ (min)
1	$32$	$30$
2	$28$	$29$
3	$41$	$37$
4	$35$	$36$
5	$24$	$26$

She wants a single number describing the typical size of a prediction error, in minutes, so she computes the root mean squared error:

$\text{RMSE} = \sqrt{\frac{1}{n}\sum_{i=1}^{n}(y_i - \hat{y}_i)^2}$

What is the RMSE of these five predictions?

Round your answer to 2 decimal places.

How Far Off Is the Delivery Model, Typically?

Hints