Analyses

Nelson–Aalen

Non-parametric cumulative hazard from duration and an event indicator.

Edit on GitHub

When to use it

Use Nelson–Aalen when you want the cumulative hazard of the same duration and event columns that Kaplan–Meier uses for survival. Survival probability is Kaplan–Meier. A covariate on the hazard is Cox.

Assumptions

Duration must be positive and numeric. The event column is 1 for the event and 0 for censored; true, yes, y, and event also count as 1. At least three valid rows are required. There is no grouping column on this procedure. The cumulative hazard chart is a step function from the origin (0, 0), matching Kaplan–Meier.

Running it in Tensr

Analyze → Survival → Nelson–Aalen. In chat: “Nelson–Aalen of weeks.”
Duration is the time column. Event is 1 = event, 0 = censored.

Options

Prop

Type

Reading the output

The same 80 durations and events, as a cumulative hazard rather than a survival curve. There is no grouping column on this procedure.

The report is the count of rows and events, not a p-value. N = 80, events = 58. The cumulative hazard chart is a step function of those events over weeks.

Reporting (APA 7)

Nelson–Aalen cumulative hazard for weeks, N = 80, 58 events. Report the curve and the event count; this estimator has no single p-value.

Coming from SPSS

SPSS does not ship a Nelson–Aalen dialog. The estimator is the Fleming–Harrington cumulative hazard from the same Surv object as Kaplan–Meier. Tensr’s Analyze menu is Survival → Nelson–Aalen. The path string stored for this item is Analyze → Survival → Nelson-Aalen.

The survival curve of the same times is Kaplan–Meier. A covariate on the hazard is Cox proportional hazards.