Fitted values are the predicted values of a response variable in a statistical model. They are computed from the predictor variables in the model, using the estimated parameters from the model fitting process.
In a simple linear regression model, for example, we might be trying to predict a dependent variable (like the price of a house) based on one or more independent variables (like the number of bedrooms, the size of the house in square feet, etc.). We create a model based on a set of data, and the model gives us a “line of best fit” that predicts the dependent variable based on the independent variables.
For any given observation in the data, the fitted value is the predicted value of the dependent variable for that observation, based on the independent variables for that observation and the model’s parameters.
Fitted values play a key role in assessing the goodness-of-fit of a model. You might look at the difference between the fitted values and the actual values (the residuals) to see how well your model fits the data. If the residuals are small and randomly scattered around zero, it suggests that your model is doing a good job of capturing the underlying relationship between the variables.