This data set gives the the number of fish species in lakes of the world; to examine the effect of the surface area of the lakes. The latitude of the lakes are also recorded.
data(fish)A data frame with 70 observations on 4 variables.
name of the lakes
number of fish species in lakes
surface area (km squared)
latitude of the lakes
This data set is also used to illustrate that the fitting algorithm can handle some larger count data.
Barbour, C. D. and Brown, J. H. (1974). Fish species diversity in lakes. The American Naturalist, 108, 473--488.
### Barbour & Brown (1974): Overdispersed Fish data
# \donttest{
data(fish)
M.fish <- glm.cmp(species ~ 1 + log(area), data = fish)
M.fish
#>
#> Call: glm.cmp(formula = species ~ 1 + log(area), data = fish)
#>
#> Linear Model Coefficients:
#> (Intercept) log(area)
#> 2.32810 0.18286
#>
#> Dispersion (nu): 0.0184
#> Degrees of Freedom: 69 Total (i.e. Null); 68 Residual
#> Null Deviance: 101.6675
#> Residual Deviance:
#> AIC: 638.8532
#>
summary(M.fish)
#>
#> Call: glm.cmp(formula = species ~ 1 + log(area), data = fish)
#>
#> Deviance Residuals:
#> Min 1Q Median 3Q Max
#> -2.1215 -0.8038 -0.3327 0.4276 2.6966
#>
#> Linear Model Coefficients:
#> Estimate Std.Err Z value Pr(>|z|)
#> (Intercept) 2.32810 0.23410 9.945 < 2e-16 ***
#> log(area) 0.18286 0.02872 6.368 1.91e-10 ***
#> ---
#> Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#>
#> (Dispersion parameter for Mean-CMP estimated to be 0.01841)
#>
#>
#> Null deviance: 101.668 on 69 degrees of freedom
#> Residual deviance: 59.451 on 68 degrees of freedom
#>
#> AIC: 638.8532
#>
# }