Calculates the mean square of the model by taking the mean of the
sum of squares between the truth, \(y\), and the predicted, \(\hat{y}\)
at each observation \(i\).
Arguments
- y
A vector of the true \(y\) values
- yhat
A vector of predicted \(\hat{y}\) values.
Value
The MSE in numeric form.
Details
The equation for MSE is:
$$\frac{1}{n}\sum\limits_{i = 1}^n {{{\left( {{y_i} - {{\hat y}_i}} \right)}^2}}$$
Examples
# Set seed for reproducibility
set.seed(100)
# Generate data
n = 1e2
y = rnorm(n)
yhat = rnorm(n, 0.5)
# Compute
o = mse(y, yhat)