PER

Probability

Poisson Event Rate Calculator

Convert an observed event count and exposure into an event rate, a per-1,000 rate, projected future events, approximate rate standard error and interval, comparison-rate ratio, and projected event cost.

Observed event countObserved exposure unitsFuture exposure unitsEntered comparison rate per 1,000Entered normal critical valueEntered cost per event

Observed rate per 1,000 exposure units-

Observed events per exposure unit-

Projected events at future exposure-

Approximate rate standard error-

Approximate lower rate per 1,000-

Approximate upper rate per 1,000-

Observed rate divided by entered comparison-

Projected event cost-

Decision view

Exposure-normalized rate and future event projection

Exposure-normalized rate and future event projectionObserved exposure establishes an intensity, the rate interval forms a risk band, and future exposure extends the expected count path.

Exact scenario comparisonFuture exposure units changes while all other entered assumptions remain constant.

Future exposure units	Observed rate per 1,000 exposure units	Observed events per exposure unit	Projected events at future exposure	Approximate rate standard error	Approximate lower rate per 1,000	Approximate upper rate per 1,000	Observed rate divided by entered comparison	Projected event cost

How to use Poisson Event Rate Calculator

Enter an event count and the exposure over which it was observed.
Use the same exposure unit for observed and future values.
Enter a comparison rate per 1,000 and a critical value.
Treat projected event count and cost as conditional on stable intensity.

Calculator guide

Understanding Poisson Event Rate Calculator

An exposure-based event rate separates how many events occurred from how much opportunity existed for those events. This calculator scales the observed rate to 1,000 exposure units, projects future events, and shows a transparent normal-approximation rate interval.

Normalize first Counts become comparable only after dividing by exposure.

Scale is presentational Per-1,000 reporting does not change intensity.

Projection is conditional Future volume assumes a stable rate.

Cost follows count Projected events multiply by cost per event.

Calculation method

How the calculation works

Estimate an exposure-based event rate, scale it to 1,000 units, project future events, and display a transparent normal-approximation rate interval. In the Poisson Event Rate Calculator, the live scenario varies future exposure units and tracks projected events at future exposure while the remaining results preserve the reconciliation path. Divide observed events by observed exposure, multiply by 1,000 for the displayed rate, and multiply the per-unit rate by future exposure. Approximate the rate standard error as sqrt(events)/exposure.

Detailed calculation process

Normalize events by exposure before projecting volume

The defaults observe 42 events over 1,200 exposure units and project the same intensity across 1,800 future units.

General formula: r = E/T; r1000 = 1000r; projected events = rF; SE(r) = sqrt(E)/T; interval1000 = 1000[r +/- c SE(r)] The count is divided by its exposure opportunity to obtain an intensity. Projection assumes that intensity remains stable when exposure changes.

What each symbol means

E Observed event count, measured as events.

T Observed exposure, such as hours, people, or miles.

r / r1000 Event rate per one and per 1,000 exposure units.

F Future exposure in the same unit as T.

c Entered normal critical value, unitless.

C Entered cost per projected event in currency per event.

Worked substitution with the default inputs

1. Calculate the observed intensity: r = 42 / 1,200 = 0.035 events per exposure unit The denominator defines the opportunity base for the rate.

2. Scale the reporting unit: r1000 = 0.035 x 1,000 = 35 events per 1,000 exposure units Scaling changes presentation, not the underlying intensity.

3. Project future events: 0.035 x 1,800 = 63 projected events Projection assumes future exposure has the same event intensity.

4. Approximate rate uncertainty: SE(r) = sqrt(42)/1,200 = 0.00540062 The Poisson count standard deviation sqrt(E) is divided by exposure.

5. Form the rate interval and cost: 1,000[0.035 +/- 1.96x0.00540062] = [24.4148, 45.5852]; 63x$250 = $15,750 The observed rate is 35/30 = 1.1667 times the entered comparison rate.

The defaults produce 35 events per 1,000 exposure units, project 63 future events and $15,750 cost, and show an approximate interval of 24.415 to 45.585 per 1,000.

Rate audit

Check whether future exposure is genuinely comparable

A precise multiplication cannot repair an unstable rate definition.

Same exposure unit Observed and future denominators must match.

Stable event definition Counting rules should not change.

Stable intensity Seasonality and mix shifts need separate modeling.

Adequate count Small counts favor exact interval methods.

Worked situations

Practical examples

Forty-two events over 1,200 units equal 35 per 1,000.
At the same rate, 1,800 future units imply 63 events.
A $250 event cost turns the projected count into $15,750.

Better inputs

Useful tips

Name the exposure unit in reports.
Check for seasonality or mix changes before projecting.
Use exact Poisson methods for small counts when appropriate.

Before relying on the result

Limitations and common mistakes

The Poisson model assumes independent events and stable intensity over exposure.
The displayed interval is a normal approximation and is weak for small counts.
Clustering, overdispersion, undercounting, changing exposure quality, and time-varying rates are not modeled.

Reference

Key terms

Exposure: Opportunity base over which events can occur.
Event rate: Observed events divided by exposure.
Poisson intensity: Expected event count per exposure unit under a stable-rate model.

Important note

Calculated directly from the entered values using the displayed formula and rounding settings.

Frequently asked questions

Why report a rate per 1,000?

It makes small per-unit intensities easier to read while preserving the same underlying rate.

Is 63 a guaranteed future count?

No. It is the expected count if the observed intensity applies to the future exposure.

Can exposure be time?

Yes, provided observed and future exposure use the same time definition.

Why can a rate interval be wide?

A limited event count creates substantial sampling variation even when exposure is large.