# deviance.tweedie.md

|                  |                 |
| ---------------- | --------------: |
| deviance.tweedie | R Documentation |

### Tweedie Deviance

#### Description

A generic S3 function to compute the *tweedie deviance* score for a\
regression model. This function dispatches to S3 methods in`deviance.tweedie()` and performs no input validation. If you supply NA\
values or vectors of unequal length (e.g. `length(x) != length(y)`), the\
underlying `C++` code may trigger undefined behavior and crash your `R`\
session.

**Defensive measures**

Because `deviance.tweedie()` operates on raw pointers, pointer-level\
faults (e.g. from NA or mismatched length) occur before any `R`-level\
error handling. Wrapping calls in `try()` or `tryCatch()` will *not*\
prevent `R`-session crashes.

To guard against this, wrap `deviance.tweedie()` in a "safe" validator\
that checks for NA values and matching length, for example:

{% code overflow="wrap" lineNumbers="true" %}

```r
safe_deviance.tweedie <- function(x, y, ...) {
  stopifnot(
    !anyNA(x), !anyNA(y),
    length(x) == length(y)
  )
  deviance.tweedie(x, y, ...)
}
```

{% endcode %}

Apply the same pattern to any custom metric functions to ensure input\
sanity before calling the underlying `C++` code.

#### Usage

```r
## Generic S3 method
## for Tweedie Deviance
deviance.tweedie(...)

## Generic S3 method
## for weighted Tweedie Deviance
weighted.deviance.tweedie(...)
```

#### Arguments

| `...` | <p>Arguments passed on to <code>deviance.tweedie.numeric</code>,<code>weighted.deviance.tweedie.numeric</code></p><p><code>actual,predicted</code></p><p>A pair of \<double> vectors of length <code>n</code>.</p><p><code>power</code></p><p>A \<double> value, default = 2. Tweedie power parameter. Either<br>power <= 0 or power >= 1.</p><p>The higher <code>power</code>, the less weight is given<br>to extreme deviations between actual and predicted values.</p><ul><li><strong>power < 0:</strong> Extreme stable distribution.<br>Requires: predicted > 0.</li><li><strong>power = 0:</strong> Normal distribution, output<br>corresponds to <code>mse()</code>, actual and predicted can be any real<br>numbers.</li><li><strong>power = 1:</strong> Poisson distribution<br>(<code>deviance.poisson()</code>). Requires: actual >= 0 and<br>predicted > 0.</li><li><strong>1 < power < 2:</strong> Compound Poisson<br>distribution. Requires: actual >= 0 and predicted > 0.</li><li><strong>power = 2:</strong> Gamma distribution<br>(<code>deviance.gamma()</code>). Requires: actual > 0 and predicted<br>> 0.</li><li><strong>power = 3:</strong> Inverse Gaussian distribution.<br>Requires: actual > 0 and predicted > 0.</li><li><strong>otherwise:</strong> Positive stable distribution.<br>Requires: actual > 0 and predicted > 0.</li></ul><p><code>w</code></p><p>A \<double> vector of sample weights.</p> |
| ----- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |

#### Value

A \<double> value

#### References

James, Gareth, et al. An introduction to statistical learning. Vol. 112.\
No. 1. New York: springer, 2013.

Hastie, Trevor. "The elements of statistical learning: data mining,\
inference, and prediction." (2009).

Virtanen, Pauli, et al. "SciPy 1.0: fundamental algorithms for\
scientific computing in Python." Nature methods 17.3 (2020): 261-272.

#### Examples

```r
## Generate actual
## and predicted values
actual_values    <- c(1.3, 0.4, 1.2, 1.4, 1.9, 1.0, 1.2)
predicted_values <- c(0.7, 0.5, 1.1, 1.2, 1.8, 1.1, 0.2)

## Evaluate performance
SLmetrics::deviance.tweedie(
   actual    = actual_values, 
   predicted = predicted_values
)
```

```

</div>

```