Residuals

Residuals

The Residuals node extracts residuals from a fitted linear model using the R residuals() function. Residuals are the differences between the observed response values and the values predicted by the model, \(e_i = y_i - \hat{y}_i\).

What it does

• Retrieves a stored linear model from the upstream Linear Model node
• Calls R’s residuals() to extract the residual values
• Returns a one-column data frame with a residuals column
• Stores the result in storr so downstream nodes (e.g. Histogram, Output CSV) can use it

How to use it

1. Fit a linear model — add a Linear Model node, enter a formula, and tick “Output to storr”
2. Connect the model — drag an edge from the Linear Model node to the Residuals node input handle
3. Click Run — residuals are extracted on the server and displayed as a table

Configuration

Setting	Required	Description
Model connection	Yes	A Linear Model node with “Output to storr” enabled
Comment	No	Annotation for generated R code

Output

Displays a table with one column:

• residuals — the residual value \(e_i = y_i - \hat{y}_i\) for each observation

The output is also stored in storr (as <model_name>_residuals) so it can be passed to downstream nodes such as a Histogram for checking normality.

Generated R code

data.frame(residuals = residuals(my_data_lm))

Tips

• The upstream Linear Model node must have “Output to storr” enabled, otherwise the model is not saved and cannot be used.
• Connect the residuals output to a Histogram or Density node to check if residuals are approximately normally distributed — a key assumption of linear regression.
• Connect to an Output CSV node to download the residuals for further analysis.