observation.error {verification}

Observation Error

Description

Quantifies observation error through use of a “Gold Standard” of observations.

Usage

observation.error(obs, gold.standard = NULL, ...) 

Arguments

obs

Observation made by method to be quantified. This information can be entered two ways. If obs is a vector of length 4, it is assumed that is contains the values c(n11, n10, n01, n00), where n11 are the number of correctly predicted events and n01 is the number of incorrectly predicted non-events.

gold.standard

The gold standard. This is considered a higher quality observation (coded {0, 1 } ).

...

Optional arguments.

Value

t

Probability of forecasting an event, when an event occurs. A perfect value is 1.

u

Probability of forecasting that no event will occur, when and event occurs. A perfect value is 0.

Note

This function is used to calculate values for the measurement.error function.

Author(s)

Matt Pocernich

See Also

measurement.error

Examples

obs <- round(runif(100))
gold <- round(runif(100) )
observation.error(obs, gold)

## Pirep gold standard

observation.error(c(43,10,17,4) )