convergeLambda: Iterates until obtaining the number of bins required so that the difference in the chosen measure (lambda, R0, life expectancy of the first bin) falls below a chosen tolerance level

Description

Increases bin number by a specified binIncrease until the difference in the chosen measure falls below a tolerance level

Usage

convergeLambda(growObj, survObj, fecObj, nBigMatrix, minSize, maxSize, 
			discreteTrans = 1, integrateType = "midpoint", correction = "none", preCensus = TRUE, tol=1e-4,binIncrease=5)
convergeR0(growObj, survObj, fecObj, nBigMatrix, minSize, maxSize, 
			discreteTrans = 1, integrateType = "midpoint", correction = "none", preCensus = TRUE, tol=1e-4,binIncrease=5)
convergeLifeExpectancyFirstBin(growObj, survObj,nBigMatrix, minSize, maxSize, 
				discreteTrans = 1, integrateType = "midpoint", correction = "none", 
				tol=1e-1,binIncrease=5)
convergeLifeExpectancyLastBin(growObj, survObj,nBigMatrix, minSize, maxSize, 
				discreteTrans = 1, integrateType = "midpoint", correction = "none", 
				tol=1e-1,binIncrease=5)

Arguments

growObj

a growth object.

survObj

a survival object.

fecObj

a fecundity object.

nBigMatrix

numeric, initial number of bins of size used in the matrix - will be increased for the assessment

minSize

numeric, minimum size used for meshpoints

maxSize

numeric, maximum size used for meshpoints of the P matrix.

discreteTrans

matrix of discrete transitions; or 1 if there is none

integrateType

integration type.

correction

correction (see createIPMPmatrix)

preCensus

boolean defining whether fecundity is pre or post census; defaults to pre

tol

desired tolerance level

binIncrease

increments in increase in the number of bins (should be an integer

Value

binIncreasethe number of bins used to increase matrix size in assessing tolerance
Pmatrixthe final Pmatrix if only LE is being considered
IPMthe final IPM
R0the final R0
lambdathe final lambda
LEthe final vector of life expectancies

Details

The pattern of change in lambda (or other response variables) can be complex, so it is advisable to start with large binIncrease and small tolerance, and then once one knows a general idea of how big the matrix needs to be, run the function again with a smaller binIncrease but start it closer to the goal.

For convergeLifeExpectancyFirstBin, if discrete stages are included via discreteTrans then convergeLifeExpectancyFirstBin will use the first discrete bin.

Examples

Run this code

dff<-generateData()
gr1<-makeGrowthObj(dff)
sv1<-makeSurvObj(dff)
fv1<-makeFecObj(dff,Transform="log")

res <- convergeLambda(growObj=gr1, survObj=sv1, fecObj=fv1, nBigMatrix=10, minSize=-2, maxSize=15, 
			discreteTrans = 1, integrateType = "midpoint", correction = "none", preCensus = TRUE, tol=1e-3,binIncrease=10)


res <- convergeR0(growObj=gr1, survObj=sv1, fecObj=fv1, nBigMatrix=10, minSize=-2, maxSize=15, 
			discreteTrans = 1, integrateType = "midpoint", correction = "none", preCensus = TRUE, tol=1e-3,binIncrease=10)

res <- convergeLifeExpectancyFirstBin(growObj=gr1, survObj=sv1, nBigMatrix=10, minSize=-2, maxSize=15, 
				discreteTrans = 1, integrateType = "midpoint", correction = "none",  tol=1e-1,binIncrease=10)

res <- convergeLifeExpectancyLastBin(growObj=gr1, survObj=sv1, nBigMatrix=10, minSize=-2, maxSize=15, 
				discreteTrans = 1, integrateType = "midpoint", correction = "none",  tol=1e-1,binIncrease=10)