RatePhylo.CI: Confidence intervals for rate parameters In ghthomas/motmot: Models of Trait Macroevolution on Trees

Description

Calculates approximate confidence intervals for all rate parameters. CIs are esimated for one rate parameters while fixing others at a given value (usually the maximum likelihood estimate).

These are reliable (given the asympotic assumptions of the chi-square distribution) if only two groups are being compared but should be regarded only as a rough approximation for =>3 different rate categories. If the rates are correlated the CIs may be underestimated.

Usage

 `1` ```RatePhylo.CI(rateData, MLrate = NULL, fixed, rateMIN = 0.001, rateMAX = 50, common.mean = FALSE, lambda.est=TRUE) ```

Arguments

 `rateData` an object of class "rateData" `MLrate` a vector of relative rate parameters. The length of the vector is equal to the number of rates being estimated. If `rate=NULL` then rates are equal. Normally these will be the maximum likelihood rate estimates. `fixed` A vector stating whether each parameter should be allowed to vary (either `FALSE` which results in a start value of 1, or a numeric start value) or should be fixed (`TRUE`). `rateMIN` Minimum value for the rate parameters `rateMAX` Maximum value for the rate parameters `common.mean` a logical specififying whether each rate category should have its own mean (`common.mean=FALSE`) or all categories should have the same mean (`common.mean=FALSE`). See Thomas et al. (2009) for a discussion on the impact of assumptions about mean on rate estimates. `lambda.est` Logical. Estimate Pagel's lambda.

Value

 `rateLci ` Lower confidence interval for rate estimate `rateUci ` Upper confidence interval for rate estimate

Author(s)

Gavin Thomas, Rob Freckleton

References

Thomas GH, Freckleton RP, & Szekely T. 2006. Comparative analyses of the influence of developmental mode on phenotypic diversification rates in shorebirds. Proceedings of the Royal Society B 273, 1619-1624.

Thomas GH, Meiri S, & Phillimore AB. 2009. Body size diversification in Anolis: novel environments and island effects. Evolution 63, 2017-2030.

Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40``` ```data(anolis.tree) data(anolis.data) ## Convert data to class rateData with a rateMatrix object as input anolis.rateMatrix <- as.rateMatrix(phy=anolis.tree, x="geo_ecomorph", data=anolis.data) anolis.rateData <- as.rateData(y="Female_SVL", x="geo_ecomorph", rateMatrix = anolis.rateMatrix, phy=NULL, data=anolis.data, log.y=TRUE) # A model with a different rate in each of the four groups. The 'fixed' command is used to determine # whether a particular rate is to be constrained or not. Use '1' to fix a group and 'FALSE' to show # that the parameter is not fixed and should be estimated. The values should be entered in the same # order as the ranking of the groups. That is, group 0 (small islands) takes position one in the # fixed vector, group 1 (large island trunk crown and trunk ground) takes position 2 and so on. The # default is to allow each group to take a different mean. #anole.ML <- optim.likRatePhylo(rateData=anolis.rateData, rate=NULL, fixed=c(FALSE,FALSE,FALSE, FALSE), #common.mean=FALSE) # ML estimates # Confidence intervals for each parameter in turn # Not run # RatePhylo.CI(rateData=anolis.rateData, MLrate = anole.ML\$MLRate, fixed=c(FALSE, TRUE, TRUE, TRUE), #rateMIN = 0.001, rateMAX = 50, common.mean = FALSE) # Not run #RatePhylo.CI(rateData=anolis.rateData, MLrate = anole.ML\$MLRate, fixed=c(TRUE, FALSE, TRUE, TRUE), #rateMIN = 0.001, rateMAX = 50, common.mean = FALSE) # Not run #RatePhylo.CI(rateData=anolis.rateData, MLrate = anole.ML\$MLRate, fixed=c(TRUE, TRUE, FALSE, TRUE), #rateMIN = 0.001, rateMAX = 50, common.mean = FALSE) # Not run #RatePhylo.CI(rateData=anolis.rateData, MLrate = anole.ML\$MLRate, fixed=c(TRUE, TRUE, FALSE, TRUE), #rateMIN = 0.001, rateMAX = 50, common.mean = FALSE) # All confidence intervals together # Not run #RatePhylo.allCI(rateData=anolis.rateData, MLrate = anole.ML\$MLRate, fixed = NULL, rateMIN = 0.001, #rateMAX = 50, common.mean = FALSE) ```

ghthomas/motmot documentation built on July 31, 2018, 12:11 a.m.