The lambda parameter

What it does

Ridge penalty λ (0 for OLS).

Empty — the lambda from Settings (0 by default); the setting applies to both STATS_LINEAR and STATS_LOGISTIC.

With lambda > 0 the model is ridge in both functions: the slopes are shrunk (not the intercept) and predictors are not checked for dependence — nothing is excluded. The linear report then gives no standard errors, t, p or VIF, the logistic one no standard errors, Wald, Sig. or odds-ratio intervals; the model F (linear) and the omnibus chi-square and AIC/BIC (logistic) are still printed but are not valid tests of a penalized model, and the report says so: the line "Calculated with lambda …" (linear) and the omnibus header (logistic).

A number of at least 0; "Infinity" or 1e309 is an error: "lambda must be a number of at least 0 (e.g. 0), got Infinity".

Values

Optional; default 0.

Should be a number of at least 0

In plain words

A teacher looked at test scores with two columns that nearly repeated each other, hours of study and practice tests. With lambda 0, the numbers came out 10 points for hours and 6 for practice tests. With lambda 0.5, they came out 8 and 5: both were pulled a little toward zero. That keeps the numbers steadier when two columns look so alike. The price is that the report no longer shows how exact each number is, or whether it could be luck, and it says so. Lambda 0 means no pulling at all, and empty means the number from Settings.

Where you use it

• STATS_LINEAR — Multiple linear regression (OLS) with hierarchical steps separated by | and interactions a*b.
• STATS_LOGISTIC — Binary logistic regression with optional ridge regularization (lambda >= 0).

Examples

• 0 — no penalty
• 0.5 — ridge

Get it in your sheet

ALS Stats is a free add-on for Google Sheets™.

Install ALS Stats from Google Workspace Marketplace (opens in a new tab)