Package 'RDS'

Description

indexing

Usage

## S3 method for class 'rds.data.frame'
x[i, j, ..., drop, warn = TRUE]

Arguments

Details

Subsetting of RDS recruitment trees does not always yield a full RDS tree. In this case, subjects whose recruiter is no longer in the dataset are considered seeds. is issued if the 'warn' parameter is TRUE. dat <- data.frame(id=c(1,2,3,4,5), recruiter.id=c(2,-1,2,-1,4), network.size.variable=c(4,8,8,2,3)) r <- as.rds.data.frame(dat) r[1:3,] # A valid pruning of the RDS tree. r[c(1,5),warn=FALSE] # recruiter.id of last row set to -1 (i.e. a seed) to maintain validity of tree

Description

indexing

Usage

## S3 replacement method for class 'rds.data.frame'
 x[i, j] <- value

Arguments

Details

Indexed assignment. If the result is not a valid rds.data.frame, an error is emitted.

Description

converts to character with minimal loss of precision for numeric variables

Usage

as.char(x, ...)

Arguments

Description

This function converts a regular R data frame into an rds.data.frame. The greatest advantage of this is that it performs integrity checks and will fail if the recruitment information in the original data frame is incomplete.

Usage

as.rds.data.frame(
  df,
  id = if (is.null(attr(df, "id"))) "id" else attr(df, "id"),
  recruiter.id = if (is.null(attr(df, "recruiter.id"))) {
     "recruiter.id"
 } else
    attr(df, "recruiter.id"),
  network.size = if (is.null(attr(df, "network.size.variable"))) {
    
    "network.size.variable"
 } else attr(df, "network.size.variable"),
  population.size = if (all(is.na(get.population.size(df, FALSE)))) {
     NULL
 } else
    get.population.size(df, FALSE),
  max.coupons = if (is.null(attr(df, "max.coupons"))) {
     NULL
 } else attr(df,
    "max.coupons"),
  notes = if (is.null(attr(df, "notes"))) {
     NULL
 } else attr(df, "time"),
  time = if (is.null(attr(df, "time"))) {
     NULL
 } else attr(df, "time"),
  check.valid = TRUE
)

Arguments

Value

An rds.data.frame object

Examples

dat <- data.frame(id=c(1,2,3,4,5), recruiter.id=c(2,-1,2,-1,4),
                  network.size.variable=c(4,8,8,2,3))
as.rds.data.frame(dat)

Description

Does various checks and throws errors if x is not a valid rds.data.frame

Usage

assert.valid.rds.data.frame(x, ...)

Arguments

Details

Throws an informative message if x is malformed.

Description

Performs a bootstrap test of independance between two categorical variables

Usage

bootstrap.contingency.test(
  rds.data,
  row.var,
  col.var,
  number.of.bootstrap.samples = 1000,
  weight.type = c("HCG", "RDS-II", "Arithmetic Mean"),
  table.only = FALSE,
  verbose = TRUE,
  ...
)

Arguments

Details

This function first estimates a Homophily Configuration Graph model for the underlying network under the assumption that the two variables are independant and that the population size is large. It then draws bootstrap RDS samples from this population distribution and calculates the chi.squared statistic on the weighted contingency table. Weights are calculated using the HCG estimator assuming a large population size.

Examples

data(faux)
bootstrap.contingency.test(rds.data=faux, row.var="X", col.var="Y",
  number.of.bootstrap.samples=50, verbose=FALSE)

Description

Calculates incidence and bootstrap confidence intervals for immunoassay data collected with RDS

Usage

bootstrap.incidence(
  rds.data,
  recent.variable,
  hiv.variable,
  N = NULL,
  weight.type = c("Gile's SS", "RDS-I", "RDS-I (DS)", "RDS-II", "Arithmetic Mean", "HCG"),
  mean.duration = 200,
  frr = 0.01,
  post.infection.cutoff = 730,
  number.of.bootstrap.samples = 1000,
  se.mean.duration = 0,
  se.frr = 0,
  confidence.level = 0.95,
  verbose = TRUE,
  ...
)

Arguments

Details

The recent.variable and hiv should be the names of logical variables. Otherwise they are converted to logical using as.numeric(x) > 0.5.

This function estimates incidence using RDS sampling wieghts. Confidence intervals are constucted using HCG bootstraps. See http://www.incidence-estimation.org/ for additional information on (non-RDS) incidence estimation.

Examples

data(faux)
faux$hiv <- faux$X == "blue"
faux$recent <- NA
faux$recent[faux$hiv] <- runif(sum(faux$hiv)) < .2
faux$recent[runif(nrow(faux)) > .5] <- NA
faux$hiv[is.na(faux$recent)][c(1,6,10,21)] <- NA
attr(faux,"time") <- "wave"
bootstrap.incidence(faux,"recent","hiv",weight.type="RDS-II", number.of.bootstrap.samples=100)

Description

Bottleneck Plot

Usage

bottleneck.plot(
  rds.data,
  outcome.variable,
  est.func = RDS.II.estimates,
  as.factor = FALSE,
  n.eval.points = 25,
  ...
)

Arguments

References

Krista J. Gile, Lisa G. Johnston, Matthew J. Salganik Diagnostics for Respondent-driven Sampling eprint arXiv:1209.6254, 2012

Examples

data(fauxmadrona)
bottleneck.plot(fauxmadrona,"disease")

Description

Compute estimates of the sampling weights of the respondent's observations based on various estimators

Usage

compute.weights(
  rds.data,
  weight.type = c("Gile's SS", "RDS-I", "RDS-I (DS)", "RDS-II", "Arithmetic Mean", "HCG"),
  N = NULL,
  subset = NULL,
  control = control.rds.estimates(),
  ...
)

Arguments

Value

A vector of weights for each of the respondents. It is of the same size as the number of rows in rds.data.

See Also

rds.I.weights, gile.ss.weights, vh.weights

Description

Utility method that overrides the standard ‘$’ list accessor to disable partial matching for ergm control.list objects

Usage

## S3 method for class 'control.list'
object$name

Arguments

Details

Executes getElement instead of $ so that element names must match exactly to be returned and partially matching names will not return the wrong object.

Value

Returns the named list element exactly matching name, or NULL if no matching elements found

Author(s)

Pavel N. Krivitsky

See Also

see getElement

Description

Auxiliary function as user interface for fine-tuning RDS.bootstrap.intervals algorithm, which computes interval estimates for via bootstrapping.

Usage

control.rds.estimates(
  confidence.level = 0.95,
  SS.infinity = 0.01,
  lowprevalence = c(8, 14),
  discrete.cutoff = 0.8,
  useC = TRUE,
  number.of.bootstrap.samples = NULL,
  hcg.reltol = sqrt(.Machine$double.eps),
  hcg.BS.reltol = 1e+05 * sqrt(.Machine$double.eps),
  hcg.max.optim = 500,
  seed = NULL
)

Arguments

Details

This function is only used within a call to the RDS.bootstrap.intervals function.

Some of the arguments are not yet fully implemented. It will evolve slower to incorporate more arguments as the package develops.

Standard confidence interval procedures can be inaccurate when the outcome expected count is close to zero. In these cases the combined Agresti-Coull and the bootstrap-t interval of Mantalos and Zografos (2008) can be used. The lowprevalence argument is a two vector parameter setting the conditions under which the approximation is used. The first is the penalty term on the differential activity. If the observed number of the rare group minus the product of the first parameter and the differential activity is lower than the second parameter, the low prevalence approximation is used.

Value

A list with arguments as components.

See Also

RDS.bootstrap.intervals

Description

This function creates diagnostic convergence plots for RDS estimators.

Usage

convergence.plot(
  rds.data,
  outcome.variable,
  est.func = RDS.II.estimates,
  as.factor = FALSE,
  n.eval.points = 25,
  ...
)

Arguments

References

Krista J. Gile, Lisa G. Johnston, Matthew J. Salganik Diagnostics for Respondent-driven Sampling eprint arXiv:1209.6254, 2012

Examples

data(faux)
convergence.plot(faux,c("X","Y"))

Description

Counts the number or recruiter->recruitee transitions between different levels of the grouping variable.

Usage

count.transitions(rds.data, group.variable)

Arguments

Examples

data(faux)
count.transitions(faux,"X")

Description

Calculates estimates at each successive wave of the sampling process

Usage

cumulative.estimate(
  rds.data,
  outcome.variable,
  est.func = RDS.II.estimates,
  n.eval.points = 25,
  ...
)

Arguments

Description

Differential Activity between groups

Usage

differential.activity.estimates(
  rds.data,
  outcome.variable,
  weight.type = "Gile's SS",
  N = NULL,
  subset = NULL,
  ...
)

Arguments

Details

This function estimates the ratio of the average degree of one population group divided by the average degree of those in another population group.

Examples

data(faux)
differential.activity.estimates(faux,"X",weight.type="RDS-II")

Description

Convert the output of print.rds.interval.estimate from a character data.frame to a numeric matrix

Usage

export.rds.interval.estimate(x, proportion = TRUE)

Arguments

Description

This is a faux set used to demonstrate RDS functions and analysis. It is used is some simple examples and has categorical variables "X", "Y" and "Z".

Format

An rds.data.frame object

References

Gile, Krista J., Handcock, Mark S., 2010 Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327.

See Also

fauxsycamore, fauxmadrona

Examples

data(faux)
RDS.I.estimates(rds.data=faux,outcome.variable='X')

Description

This is a faux set used to illustrate how the estimators perform under different populations and RDS schemes.

Format

An rds.data.frame

Details

The population had N=1000 nodes. In this case, the sample size is 500 so that there is a relatively small sample fraction (50%). There is homophily on disease status (R=5) and there is differential activity by disease status whereby the infected nodes have mean degree twice that of the uninfected (w=1.8).

In the sampling, the seeds are chosen randomly from the full population, so there is no dependency induced by seed selection.

Each sample member is given 2 uniquely identified coupons to distribute to other members of the target population in their acquaintance. Further each respondent distributes their coupons completely at random from among those they are connected to.

Here are the results for this data set and the sister fauxsycamore data set:

Even with only 50% sample, the VH is substantially biased , and the SS does much better.

Source

The original network is included as fauxmadrona.network as a network object.
The data set also includes the data.frame of the RDS data set as fauxmadrona.
Use data(package="RDS") to get a full list of datasets.

References

Gile, Krista J., Handcock, Mark S., 2010 Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327.

See Also

fauxsycamore, faux

Description

This is a faux set used to demonstrate RDS functions and analysis. The population had N=715 nodes. In this case, the sample size is 500 so that there is a relatively large sample fraction (70%). There is homophily on disease status (R=5) and there is differential activity by disease status whereby the infected nodes have mean degree twice that of the uninfected (w=1.8).

Format

An rds.data.frame plus the original network as a network object

Details

In the sampling the seeds are chosen randomly from the infected population, so there is extreme dependency induced by seed selection.

Each sample member is given 2 uniquely identified coupons to distribute to other members of the target population in their acquaintance. Further each respondent distributes their coupons completely at random from among those they are connected to.

With 70% sample, the VH is substantially biased, so the SS (and presumably MA) do much better. We expect the MA to perform a bit better than the SS.

It is network 702 and its sample from YesYes on mosix. Look for "extract702.R"
The original network is included as fauxsycamore.network as a network object.
The data set also includes the data.frame of the RDS data set as fauxsycamore.
Use data(package="RDS") to get a full list of datasets.

References

Gile, Krista J., Handcock, Mark S., 2009. Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327.

See Also

faux, fauxmadrona

Description

This is a faux set used to demonstrate RDS functions and analysis.

Format

An rds.data.frame object

References

Gile, Krista J., Handcock, Mark S., 2010 Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327.

See Also

fauxsycamore, fauxmadrona

Description

Get Horvitz-Thompson estimator assuming inclusion probability proportional to the inverse of network.var (i.e. degree).

Usage

get.h.hat(
  rds.data,
  group.variable,
  network.var = attr(rds.data, "network.size")
)

Arguments

Description

Get the subject id

Usage

get.id(x, check.type = TRUE)

Arguments

Details

returns the variable indicated by the 'id' attribute, coercing to a character vector

Description

Returns the network size of each subject (i.e. their degree).

Usage

get.net.size(x, check.type = TRUE)

Arguments

Description

Calculates the number of (direct) recuits for each respondent.

Usage

get.number.of.recruits(data)

Arguments

Examples

data(fauxmadrona)
nr <- get.number.of.recruits(fauxmadrona)
#frequency of number recruited by each id
barplot(table(nr))

Description

Returns the population size associated with the data.

Usage

get.population.size(x, check.type = TRUE)

Arguments

Description

Returns the recruitment time for each subject

Usage

get.recruitment.time(
  x,
  to.numeric = TRUE,
  wave.fallback = FALSE,
  check.type = TRUE
)

Arguments

Description

Get recruiter id

Usage

get.rid(x, check.type = TRUE)

Arguments

Details

returns the variable indicated by the 'recruiter.id' attribute, coercing to a character vector

Description

Calculates the root seed id for each node of the recruitement tree.

Usage

get.seed.id(data)

Arguments

Examples

data(fauxmadrona)
seeds <- get.seed.id(fauxmadrona)
#number recruited by each seed
barplot(table(seeds))

Description

Gets the recruiter id associated with the seeds

Usage

get.seed.rid(x, check.type = TRUE)

Arguments

Details

All seed nodes must have the same placeholder recruiter id.

Description

Markov chain statistionary distribution

Usage

get.stationary.distribution(mle)

Arguments

Value

A vector of proportions representing the proportion in each group at the stationary distribution of the Markov chain.

Description

Calculates the depth of the recruitment tree (i.e. the recruitment wave) at each node.

Usage

get.wave(data)

Arguments

Examples

data(fauxmadrona)
#number subjects in each wave
w <- get.wave(fauxmadrona)
#number recruited in each wave
barplot(table(w))

Description

Weights using Giles SS estimator

Usage

gile.ss.weights(
  degs,
  N,
  number.ss.samples.per.iteration = 500,
  number.ss.iterations = 5,
  hajek = TRUE,
  SS.infinity = 0.04,
  se = FALSE,
  ...
)

Arguments

Description

RDS data.frame has recruitment time information

Usage

has.recruitment.time(x, check.type = TRUE)

Arguments

Description

HCG parametric bootstrap replicate weights

Usage

hcg.replicate.weights(
  rds.data,
  outcome.variable,
  number.of.bootstrap.samples = 500,
  include.sample.weights = FALSE,
  N = NULL,
  small.fraction = FALSE
)

Arguments

Details

This function generates bootstrap replicate weights which may be used to analyze RDS data in other packages or software systems (e.g. the survey package with svrepdesign).

Value

A data.frame of replicate weights. If include.sample.weights is TRUE, the first column are the HCG weights for the observed sample.

Examples

## Not run: 
data("fauxmadrona")
set.seed(1)
# Generate replicate weights
result <- hcg.replicate.weights(fauxmadrona, "disease", 50, TRUE)
# Analyze with survey package and compare to internal function
if(require(survey)){
  set.seed(1)
  design <- svrepdesign(fauxmadrona, type = "bootstrap", 
    weights= result[[1]], repweights = result[-1])
  svymean(~disease, design) |> print()
  RDS.bootstrap.intervals(fauxmadrona, "disease", "HCG", "HCG", 
  number.of.bootstrap.samples = 50) |> print()
}

## End(Not run)

Description

homophily configuration graph weights

Usage

hcg.weights(
  rds.data,
  outcome.variable,
  N = NULL,
  small.fraction = FALSE,
  reltol = sqrt(.Machine$double.eps),
  max.optim = 500,
  theta.start = NULL,
  weights.include.seeds = TRUE,
  ...
)

Arguments

Examples

data(fauxtime)
hcg.weights(fauxtime,"var1",N=3000)
fauxtime$NETWORK[c(1,100,40,82,77)] <- NA

Description

This function computes an estimate of the population homophily and the recruitment homophily based on a categorical variable.

Usage

homophily.estimates(
  rds.data,
  outcome.variable,
  weight.type = NULL,
  uncertainty = NULL,
  recruitment = FALSE,
  N = NULL,
  to.group0.variable = NULL,
  to.group1.variable = NULL,
  number.ss.samples.per.iteration = NULL,
  confidence.level = 0.95
)

Arguments

Value

If outcome.variable is binary then the homophily estimate of 0 verses 1 is returned, otherwise a vector of differential homophily estimates is returned.

Recruitment Homophily

The recruitment homophily is a homophily measure for the recruitment process. It addresses the question: Do respondents differential recruit people like themselves? That is, the homophily on a variable in the recruitment chains. Take as an example infection status. In this case, it is the ratio of number of recruits that have the same infection status as their recruiter to the number we would expect if there was no homophily on infection status. The difference with the Population Homophily (see below) is that this is in the recruitment chain rather than the population of social ties. For example, of the recruitment homophily on infection status is about 1, we see little effect of recruitment homophily on infection status (as the numbers of homophilous pairs are close to what we would expect by chance).

Population Homophily

This is an estimate the homophily of a given variable in the underlying networked population. For example, consider HIV status. The population homophily is the homophily in the HIV status of two people who are tied in the underlying population social network (a “couple”). Specifically, the population homophily is the ratio of the expected number of HIV discordant couples absent homophily to the expected number of HIV discordant couples with the homophily. Hence larger values of population homophily indicate more homophily on HIV status. For example, a value of 1 means the couple are random with respect to HIV status. A value of 2 means there are twice as many HIV discordant couples as we would expect if there was no homophily in the population. This measure is meaningful across different levels of differential activity. As we do not see most of the population network, we estimate the population homophily from the RDS data. As an example, suppose the population homophily on HIV is 0.75 so there are 25% more HIV discordant couples than expected due to chance. So their is actually heterophily on HIV in the population. If the population homophily on sex is 1.1, there are 10% more same-sex couples than expected due to chance. Hence there is modest homophily on sex.

Author(s)

Mark S. Handcock with help from Krista J. Gile

References

Gile, Krista J., Handcock, Mark S., 2010, Respondent-driven Sampling: An Assessment of Current Methodology. Sociological Methodology 40, 285-327.

Examples

## Not run: 
data(fauxmadrona)
names(fauxmadrona)
#
# True value:
#
if(require(network)){
	a=as.sociomatrix(fauxmadrona.network)
	deg <- apply(a,1,sum)
	dis <- fauxmadrona.network \
	deg1 <- apply(a[dis==1,],1,sum)
	deg0 <- apply(a[dis==0,],1,sum)
	# differential activity
	mean(deg1)/ mean(deg0)
	p=mean(dis)
	N=1000
	# True homophily
	p*(1-p)*mean(deg0)*mean(deg1)*N/(mean(deg)*sum(a[dis==1,dis==0]))
}
# HT based estimators using the to.group information
data(fauxmadrona)
homophily.estimates(fauxmadrona,outcome.variable="disease",
  to.group0.variable="tonondiseased", to.group1.variable="todiseased",
  N=1000)
# HT based estimators not using the to.group information
homophily.estimates(fauxmadrona,outcome.variable="disease",
  N=1000,weight.type="RDS-II")

## End(Not run)

Description

Imputes missing degree values

Usage

impute.degree(
  rds.data,
  trait.variable = NULL,
  N = NULL,
  method = c("mean", "quantile"),
  quantile = 0.5,
  recruitment.lower.bound = TRUE,
  round.degree = TRUE
)

Arguments

Details

This function imputes degree values using the weighted mean or quantile values of the non-missing degrees. Weights are calcualted using Gile's SS if N is not NULL, or RDS-II if it is. If a trait variable is specified, means and quantile are calculated within the levels of the trait variable

Examples

data(faux)
rds.data <- faux
rds.data$network.size[c(1,2,30,52,81,101,108,111)] <- NA
impute.degree(rds.data)
impute.degree(rds.data,trait.variable="X")
impute.degree(rds.data,trait.variable="X",method="quantile")

Description

Estimates each person's personal visibility based on their self-reported degree and the number of their (direct) recruits. It uses the time the person was recruited as a factor in determining the number of recruits they produce.

Usage

impute.visibility(
  rds.data,
  max.coupons = NULL,
  type.impute = c("median", "distribution", "mode", "mean"),
  recruit.time = NULL,
  include.tree = FALSE,
  reflect.time = FALSE,
  parallel = 1,
  parallel.type = "PSOCK",
  interval = 10,
  burnin = 5000,
  mem.optimism.prior = NULL,
  df.mem.optimism.prior = 5,
  mem.scale.prior = 2,
  df.mem.scale.prior = 10,
  mem.overdispersion = 15,
  return.posterior.sample.visibilities = FALSE,
  verbose = FALSE
)

Arguments

References

McLaughlin, Katherine R.; Johnston, Lisa G.; Jakupi, Xhevat; Gexha-Bunjaku, Dafina; Deva, Edona and Handcock, Mark S. (2023) Modeling the Visibility Distribution for Respondent-Driven Sampling with Application to Population Size Estimation, Annals of Applied Statistics, doi:10.1093/jrsssa/qnad031

Examples

## Not run: 
data(fauxmadrona)
# The next line fits the model for the self-reported personal
# network sizes and imputes the personal network sizes 
# It may take up to 60 seconds.
visibility <- impute.visibility(fauxmadrona)
# frequency of estimated personal visibility
table(visibility)

## End(Not run)

Description

Estimates each person's personal visibility based on their self-reported degree and the number of their (direct) recruits. It uses the time the person was recruited as a factor in determining the number of recruits they produce.

Usage

impute.visibility_mle(
  rds.data,
  max.coupons = NULL,
  type.impute = c("distribution", "mode", "median", "mean"),
  recruit.time = NULL,
  include.tree = FALSE,
  unit.scale = NULL,
  unit.model = c("cmp", "nbinom"),
  optimism = FALSE,
  guess = NULL,
  reflect.time = TRUE,
  maxit = 100,
  K = NULL,
  verbose = TRUE
)

Arguments

References

McLaughlin, K.R., M.S. Handcock, and L.G. Johnston, 2015. Inference for the visibility distribution for respondent-driven sampling. In JSM Proceedings. Alexandria, VA: American Statistical Association. 2259-2267.

Examples

## Not run: 
data(fauxmadrona)
# The next line fits the model for the self-reported personal
# network sizes and imputes the personal network sizes 
# It may take up to 60 seconds.
visibility <- impute.visibility(fauxmadrona)
# frequency of estimated personal visibility
table(visibility)

## End(Not run)

Description

Is an instance of rds.data.frame

Usage

is.rds.data.frame(x)

Arguments

Description

Is an instance of rds.interval.estimate

Usage

is.rds.interval.estimate(x)

Arguments

Description

Is an instance of rds.interval.estimate.list This is a (typically time ordered) sequence of RDS estimates of a comparable quantity

Usage

is.rds.interval.estimate.list(x)

Arguments

Description

This function takes a series of point estimates and their associated standard errors and computes the p-value for the test of a monotone decrease in the population prevalences (in sequence order). The p-value for a monotone increase is also reported. An optional plot of the estimates and the null distribution of the test statistics is provided. More formally, let the $K$ population prevalences in sequence order be $p_1, \ldots, p_K$. We test the null hypothesis:

$H_0 : p_1 = \ldots = p_K$

vs

$H_1 : p_1 \ge p_2 \ldots \ge p_K$

with at least one equality strict. The alternatie hypothesis is for a monotone decreasing trend. A likelihood ratio statistic for this test has been derived (Bartholomew 1959). The null distribution of the likelihood ratio statistic is very complex but can be determined by a simple Monte Carlo process.
Alternatively, we can test the null hypothesis:

$H_0 : p_1 \ge p_2 \ldots \ge p_K$

vs

$H_1 : \overline{H_0}$

The null distribution of the likelihood ratio statistic is very complex but can be determined by a simple Monte Carlo process. In both cases we also test for:

$H : p_1 \le p_2 \ldots \le p_K$

that is, a monotonically increasing trend. The function requires the isotone library.

Usage

LRT.trend.test(
  data,
  variables = colnames(data),
  null = "monotone",
  confidence.level = 0.95,
  number.of.bootstrap.samples = 5000,
  plot = NULL,
  seed = 1
)

Arguments

Value

A list with components

• pvalue.increasing: The p-value for the test of a monotone increase in population prevalence.

• pvalue.decreasing: The p-value for the test of a monotone decrease in population prevalence.

• L: The value of the likelihood-ratio statistic.

• x: The passed vector of prevalence estimates in the order (e.g., time).

• sigma The passed vector of standard error estimates corresponding to x.

Author(s)

Mark S. Handcock

References

Bartholomew, D. J. (1959). A test of homogeneity for ordered alternatives. Biometrika 46 36-48.

Examples

d <- t(data.frame(estimate=c(0.16,0.15,0.3), sigma=c(0.04,0.04,0.1)))
colnames(d) <- c("time_1","time_2","time_3") 
LRT.trend.test(d,number.of.bootstrap.samples=1000)

Description

This function takes a series of point estimates and their associated standard errors and computes the p-value for the test of a monotone decrease in the population prevalences (in sequence order). The p-value for a monotone increase is also reported. More formally, let the $K$ population prevalences in sequence order be $p_1, \ldots, p_K$. We test the null hypothesis:

$H_0 : p_1 = \ldots = p_K$

vs

$H_1 : p_1 \ge p_2 \ldots \ge p_K$

with at least one equality strict. A likelihood ratio statistic for this test has been derived (Bartholomew 1959). The null distribution of the likelihood ratio statistic is very complex but can be determined by a simple Monte Carlo process.
We also test the null hypothesis:

$H_0 : p_1 \ge p_2 \ldots \ge p_K$

vs

$H_1 : \overline{H_0}$

The null distribution of the likelihood ratio statistic is very complex but can be determined by a simple Monte Carlo process. The function requires the isotone library.

Usage

LRT.value.trend(x, sigma)

Arguments

Value

A list with components

• pvalue.increasing: The p-value for the test of a monotone increase in population prevalence.

• pvalue.decreasing: The p-value for the test of a monotone decrease in population prevalence.

• L: The value of the likelihood-ratio statistic.

• x: The passed vector of prevalence estimates in the order (e.g., time).

• sigma The passed vector of standard error estimates corresponding to x.

Author(s)

Mark S. Handcock

References

Bartholomew, D. J. (1959). A test of homogeneity for ordered alternatives. Biometrika 46 36-48.

Examples

## Not run: 
x <- c(0.16,0.15,0.3)
sigma <- c(0.04,0.04,0.1)
LRT.value.trend(x,sigma)

## End(Not run)

Description

This function computes the sequential sampling (MA) estimates for a categorical variable or numeric variable.

Usage

MA.estimates(
  rds.data,
  trait.variable,
  seed.selection = "degree",
  number.of.seeds = NULL,
  number.of.coupons = NULL,
  number.of.iterations = 3,
  N = NULL,
  M1 = 25,
  M2 = 20,
  seed = 1,
  initial.sampling.probabilities = NULL,
  MPLE.samplesize = 50000,
  SAN.maxit = 5,
  SAN.nsteps = 2^19,
  sim.interval = 10000,
  number.of.cross.ties = NULL,
  max.degree = NULL,
  parallel = 1,
  parallel.type = "PSOCK",
  full.output = FALSE,
  verbose = TRUE
)

Arguments

Value

If trait.variable is numeric then the model-assisted estimate of the mean is returned, otherwise a vector of proportion estimates is returned. If full.output=TRUE this leads to:

If full.output=FALSE this leads to an object of class rds.interval.estimate which is a list with component

estimate

the numerical point estimate of proportion of thetrait.variable.

interval

a matrix with size columns and one row per category of trait.variable:

point estimate

The HT estimate of the population mean.

95% Lower Bound

Lower 95% confidence bound

95% Upper Bound

Upper 95% confidence bound

Design Effect

The design effect of the RDS

s.e.

standard error

n

count of the number of sample values with that value of the trait

rds.data

an rds.data.frame that indicates recruitment patterns by a pair of attributes named “id” and “recruiter.id”.

N

an estimate of the number of members of the population being sampled. If NULL it is read as the pop.size.mid attribute of the rds.data frame. If that is missing it defaults to 1000.

M1

the number of networked populations generated at each iteration.

M2

the number of (full) RDS populations generated for each networked population at each iteration.

seed

the random number seed used to initiate the computations.

seed.selection

an estimate of the mechanism guiding the choice of seeds. The choices are

"allwithtrait"

indicating that all the seeds had the trait;

"random"

meaning they were, as if, a simple random sample of individuals from the population;

"sample"

indicating that the seeds are taken as those in the sample (and resampled for the population with that composition if necessary);

"degree"

is proportional to the degree of the individual;

"allwithtraitdegree"

indicating that all the seeds had the trait and the probability of being a seed is proportional to the degree of the respondent.

number.of.seeds

The number of seeds chosen to initiate the sampling.

number.of.coupons

The number of coupons given to each respondent.

number.of.iterations

The number of iterations used at the core of the algorithm.

outcome.variable

The name of the outcome variable

weight.type

The type of weighting used (i.e. MA)

uncertainty

The type of weighting used (i.e. MA)

details

A list of other diagnostic output from the computations.

varestBS

Output from the bootstrap procedure. A list with two elements: var is the bootstrap variance, and BSest is the vector of bootstrap estimates themselves.

coefficient

estimate of the parameter of the ERGM for the network.

Author(s)

Krista J. Gile with help from Mark S. Handcock

References

Gile, Krista J. 2011 Improved Inference for Respondent-Driven Sampling Data with Application to HIV Prevalence Estimation, Journal of the American Statistical Association, 106, 135-146.

Gile, Krista J., Handcock, Mark S., 2010. Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327. <doi:10.1111/j.1467-9531.2010.01223.x>

Gile, Krista J., Beaudry, Isabelle S. and Handcock, Mark S., 2018 Methods for Inference from Respondent-Driven Sampling Data, Annual Review of Statistics and Its Application <doi:10.1146/annurev-statistics-031017-100704>.

See Also

RDS.I.estimates, RDS.I.estimates

Examples

## Not run: 
data(faux)
MA.estimates(rds.data=faux,trait.variable='X')

## End(Not run)

Description

Diagnostic plots for the RDS recruitment process

Usage

## S3 method for class 'rds.data.frame'
plot(
  x,
  plot.type = c("Recruitment tree", "Network size by wave", "Recruits by wave",
    "Recruits per seed", "Recruits per subject"),
  stratify.by = NULL,
  ...
)

Arguments

Details

Several types of diagnostics are supported by the plot.type argument. 'Recruitment tree' displays a network plot of the RDS recruitment process. 'Network size by wave' monitors systematic changes is network size based on how far subjects are from the seed 'Recruits by wave' displays counts of subjects based on how far they rare from their seed. 'Recruit per seed' shows the total tree size for each seed. 'Recruits per subject' shows counts of how many subjects are recruited by each subject who are non-terminal.

Value

Either nothing (for the recruitment tree plot), or a ggplot2 object.

Examples

data(fauxmadrona)
## Not run: 
plot(fauxmadrona)

## End(Not run)
plot(fauxmadrona, plot.type='Recruits by wave')
plot(fauxmadrona, plot.type='Recruits per seed')
plot(fauxmadrona, plot.type='Recruits per subject')

plot(fauxmadrona, plot.type='Recruits by wave', stratify.by='disease')
plot(fauxmadrona, plot.type='Recruits per seed', stratify.by='disease')
plot(fauxmadrona, plot.type='Recruits per subject', stratify.by='disease')

Description

Prints an differential.activity.estimate object

Usage

## S3 method for class 'differential.activity.estimate'
print(x, ...)

Arguments

Description

Displays a pvalue.table

Usage

## S3 method for class 'pvalue.table'
print(x, ...)

Arguments

Description

Displays an rds.contin.bootstrap

Usage

## S3 method for class 'rds.contin.bootstrap'
print(x, show.table = FALSE, ...)

Arguments

Description

Displays an rds.data.frame

Usage

## S3 method for class 'rds.data.frame'
print(x, ...)

Arguments

Description

Prints an rds.interval.estimate object

Usage

## S3 method for class 'rds.interval.estimate'
print(x, as.percentage = NULL, ...)

Arguments

Description

print.summary.svyglm.RDS is a version of print.summary.svyglm that reports odds-ratios in place of coefficients in the summary table. This only applies for the binomial family. Otherwise it is identical to print.summary.svyglm. The default in
print.summary.svyglm is to display the log-odds-ratios and this displays the exponetiated from and a 95 p-values are still displayed.

Usage

## S3 method for class 'summary.svyglm.RDS'
print(
  x,
  digits = max(3, getOption("digits") - 3),
  symbolic.cor = x$symbolic.cor,
  signif.stars = getOption("show.signif.stars"),
  ...
)

Arguments

See Also

svyglm, summary.svyglm.

Examples

## For examples see example(svyglm)

Description

This function computes an interval estimate for one or more categorical variables. It optionally uses attributes of the RDS data set to determine the type of estimator and type of uncertainty estimate to use.

Usage

RDS.bootstrap.intervals(
  rds.data,
  outcome.variable,
  weight.type = NULL,
  uncertainty = NULL,
  N = NULL,
  subset = NULL,
  confidence.level = 0.95,
  number.of.bootstrap.samples = NULL,
  fast = TRUE,
  useC = TRUE,
  ci.type = "t",
  control = control.rds.estimates(),
  to.factor = FALSE,
  cont.breaks = 3,
  ...
)

Arguments

Value

An object of class rds.interval.estimate summarizing the inference. The confidence interval and standard error are based on the bootstrap procedure. In additon, the object has attribute bsresult which provides details of the bootstrap procedure. The contents of the bsresult attribute depends on the uncertainty used. If uncertainty=="Salganik" then bsresult is a vector of standard deviations of the bootstrap samples. If uncertainty=="Gile's SS" then bsresult is a list with components for the bootstrap point estimate, the bootstrap samples themselves and the standard deviations of the bootstrap samples. If uncertainty=="SRS" then bsresult is NULL.

References

Gile, Krista J. 2011 Improved Inference for Respondent-Driven Sampling Data with Application to HIV Prevalence Estimation, Journal of the American Statistical Association, 106, 135-146.

Gile, Krista J., Handcock, Mark S., 2010. Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327. <doi:10.1111/j.1467-9531.2010.01223.x>

Gile, Krista J., Beaudry, Isabelle S. and Handcock, Mark S., 2018 Methods for Inference from Respondent-Driven Sampling Data, Annual Review of Statistics and Its Application <doi:10.1146/annurev-statistics-031017-100704>.

Examples

## Not run: 
data(fauxmadrona)
RDS.bootstrap.intervals(rds.data=fauxmadrona,weight.type="RDS-II",
     uncertainty="Salganik",
	outcome.variable="disease",N=1000,number.of.bootstrap.samples=50)

data(fauxtime)
RDS.bootstrap.intervals(rds.data=fauxtime,weight.type="HCG",
     uncertainty="HCG",
	outcome.variable="var1",N=1000,number.of.bootstrap.samples=10)

## End(Not run)

Description

Compares the rates of two variables against one another.

Usage

RDS.compare.proportions(first.interval, second.interval, M = 10000)

Arguments

Details

This function preforms a bootstrap test comparing the the rates of two variables against one another.

Examples

## Not run: 
data(faux)
int1 <- RDS.bootstrap.intervals(faux, outcome.variable=c("X"), 
weight.type="RDS-II", uncertainty="Salganik", N=1000,
number.ss.samples.per.iteration=1000, 
	confidence.level=0.95, number.of.bootstrap.samples=100)
int2 <- RDS.bootstrap.intervals(faux, outcome.variable=c("Y"), 
	weight.type="RDS-II", uncertainty="Salganik", N=1000,
number.ss.samples.per.iteration=1000,
confidence.level=0.95, number.of.bootstrap.samples=100)
RDS.compare.proportions(int1,int2)

## End(Not run)

Description

Compares the rates of two variables against one another.

Usage

RDS.compare.two.proportions(
  data,
  variables,
  confidence.level = 0.95,
  number.of.bootstrap.samples = 5000,
  plot = FALSE,
  seed = 1
)

Arguments

Value

An object of class pvalue.table containing the cross-tabulation of p-values for comparing the two classes

Description

This function computes the Homophily Configuration Graph type estimates for a categorical variable.

Usage

RDS.HCG.estimates(
  rds.data,
  outcome.variable,
  N = NULL,
  subset = NULL,
  small.fraction = FALSE,
  empir.lik = TRUE,
  to.factor = FALSE,
  cont.breaks = 3
)

Arguments

Value

If the empir.lik is true, an object of class rds.interval.estimate is returned. This is a list with components

• estimate: The numerical point estimate of proportion of the trait.variable.

• interval: A matrix with six columns and one row per category of trait.variable:

  • point estimate: The HT estimate of the population mean.

  • 95% Lower Bound: Lower 95% confidence bound.

  • 95% Upper Bound: Upper 95% confidence bound.

  • Design Effect: The design effect of the RDS.

  • s.e.: Standard error.

  • n: Count of the number of sample values with that value of the trait.

Otherwise an object of class rds.HCG.estimate object is returned.

Author(s)

Ian E. Fellows

See Also

RDS.I.estimates, RDS.II.estimates, RDS.SS.estimates

Examples

data(fauxtime)
RDS.HCG.estimates(rds.data=fauxtime,outcome.variable='var1')

Description

This function computes the RDS-I type estimates for a categorical variable. It is also referred to as the Salganik-Heckathorn estimator.

Usage

RDS.I.estimates(
  rds.data,
  outcome.variable,
  N = NULL,
  subset = NULL,
  smoothed = FALSE,
  empir.lik = TRUE,
  to.factor = FALSE,
  cont.breaks = 3
)

Arguments

Value

If the empir.lik is true, an object of class rds.interval.estimate is returned. This is a list with components

• estimate: The numerical point estimate of proportion of the trait.variable.

• interval: A matrix with six columns and one row per category of trait.variable:

  • point estimate: The HT estimate of the population mean.

  • 95% Lower Bound: Lower 95% confidence bound.

  • 95% Upper Bound: Upper 95% confidence bound.

  • Design Effect: The design effect of the RDS.

  • s.e.: Standard error.

  • n: Count of the number of sample values with that value of the trait.

Otherwise an object of class rds.I.estimate object is returned.

Author(s)

Mark S. Handcock and W. Whipple Neely

References

Gile, Krista J., Handcock, Mark S., 2010. Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327. <doi:10.1111/j.1467-9531.2010.01223.x>

Gile, Krista J., Beaudry, Isabelle S. and Handcock, Mark S., 2018 Methods for Inference from Respondent-Driven Sampling Data, Annual Review of Statistics and Its Application <doi:10.1146/annurev-statistics-031017-100704>.

Neely, W. W., 2009. Bayesian methods for data from respondent driven sampling. Dissertation in-progress, Department of Statistics, University of Wisconsin, Madison.

Salganik, M., Heckathorn, D. D., 2004. Sampling and estimation in hidden populations using respondent-driven sampling. Sociological Methodology 34, 193-239.

Volz, E., Heckathorn, D., 2008. Probability based estimation theory for Respondent Driven Sampling. The Journal of Official Statistics 24 (1), 79-97.

See Also

RDS.II.estimates, RDS.SS.estimates

Examples

data(faux)
RDS.I.estimates(rds.data=faux,outcome.variable='X')
RDS.I.estimates(rds.data=faux,outcome.variable='X',smoothed=TRUE)

Description

RDS-I weights

Usage

rds.I.weights(rds.data, outcome.variable, N = NULL, smoothed = FALSE, ...)

Arguments

Description

This function computes the RDS-II estimates for a categorical variable or the RDS-II estimate for a numeric variable.

Usage

RDS.II.estimates(
  rds.data,
  outcome.variable,
  N = NULL,
  subset = NULL,
  empir.lik = TRUE,
  to.factor = FALSE
)

Arguments

Value

If outcome.variable is numeric then the RDS-II estimate of the mean is returned, otherwise a vector of proportion estimates is returned. If the empir.lik is true, an object of class rds.interval.estimate is returned. This is a list with components

• estimate: The numerical point estimate of proportion of the trait.variable.

• interval: A matrix with six columns and one row per category of trait.variable:

  • point estimate: The HT estimate of the population mean.

  • 95% Lower Bound: Lower 95% confidence bound.

  • 95% Upper Bound: Upper 95% confidence bound.

  • Design Effect: The design effect of the RDS.

  • s.e.: Standard error.

  • n: Count of the number of sample values with that value of the trait.

Otherwise, an object of class rds.II.estimate is returned.

Author(s)

Mark S. Handcock and W. Whipple Neely

References

Gile, Krista J., Handcock, Mark S., 2010. Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327. <doi:10.1111/j.1467-9531.2010.01223.x>

Gile, Krista J., Beaudry, Isabelle S. and Handcock, Mark S., 2018 Methods for Inference from Respondent-Driven Sampling Data, Annual Review of Statistics and Its Application <doi:10.1146/annurev-statistics-031017-100704>.

Salganik, M., Heckathorn, D. D., 2004. Sampling and estimation in hidden populations using respondent-driven sampling. Sociological Methodology 34, 193-239.

Volz, E., Heckathorn, D., 2008. Probability based estimation theory for Respondent Driven Sampling. The Journal of Official Statistics 24 (1), 79-97.

See Also

RDS.I.estimates, RDS.SS.estimates

Examples

data(faux)
RDS.II.estimates(rds.data=faux,outcome.variable='X')
RDS.II.estimates(rds.data=faux,outcome.variable='X',subset= Y!="blue")

Description

This function creates an object of class rds.interval.estimate.

Usage

rds.interval.estimate(
  estimate,
  outcome.variable,
  weight.type,
  uncertainty,
  weights,
  N = NULL,
  conf.level = 0.95,
  csubset = ""
)

Arguments

Value

An object of class rds.interval.estimate is returned. This is a list with components

• estimate: The numerical point estimate of proportion of the trait.variable.

• interval: A matrix with six columns and one row per category of trait.variable:

  • point estimate: The HT estimate of the population mean.

  • 95% Lower Bound: Lower 95% confidence bound.

  • 95% Upper Bound: Upper 95% confidence bound.

  • Design Effect: The design effect of the RDS.

  • s.e.: Standard error.

  • n: Count of the number of sample values with that value of the trait.

Author(s)

Mark S. Handcock

References

Gile, Krista J., Handcock, Mark S., 2010. Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327. <doi:10.1111/j.1467-9531.2010.01223.x>

Gile, Krista J., Beaudry, Isabelle S. and Handcock, Mark S., 2018 Methods for Inference from Respondent-Driven Sampling Data, Annual Review of Statistics and Its Application <doi:10.1146/annurev-statistics-031017-100704>.

Salganik, M., Heckathorn, D. D., 2004. Sampling and estimation in hidden populations using respondent-driven sampling. Sociological Methodology 34, 193-239.

Volz, E., Heckathorn, D., 2008. Probability based estimation theory for Respondent Driven Sampling. The Journal of Official Statistics 24 (1), 79-97.

Examples

data(faux)
RDS.I.estimates(rds.data=faux,outcome.variable='X',smoothed=TRUE)

Description

This function computes the sequential sampling (SS) estimates for a categorical variable or numeric variable.

Usage

RDS.SS.estimates(
  rds.data,
  outcome.variable,
  N = NULL,
  subset = NULL,
  number.ss.samples.per.iteration = 500,
  number.ss.iterations = 5,
  control = control.rds.estimates(),
  hajek = TRUE,
  empir.lik = TRUE,
  to.factor = FALSE
)

Arguments

Value

If outcome.variable is numeric then the Gile SS estimate of the mean is returned, otherwise a vector of proportion estimates is returned. If the empir.lik is true, an object of class rds.interval.estimate is returned. This is a list with components

• estimate: The numerical point estimate of proportion of the trait.variable.

• interval: A matrix with six columns and one row per category of trait.variable:

  • point estimate: The HT estimate of the population mean.

  • 95% Lower Bound: Lower 95% confidence bound.

  • 95% Upper Bound: Upper 95% confidence bound.

  • Design Effect: The design effect of the RDS.

  • s.e.: Standard error.

  • n: Count of the number of sample values with that value of the trait.

Otherwise, an object of class rds.SS.estimate is returned.

Author(s)

Krista J. Gile with help from Mark S. Handcock

References

Gile, Krista J. 2011 Improved Inference for Respondent-Driven Sampling Data with Application to HIV Prevalence Estimation, Journal of the American Statistical Association, 106, 135-146.

Gile, Krista J., Handcock, Mark S., 2010. Respondent-driven Sampling: An Assessment of Current Methodology, Sociological Methodology, 40, 285-327. <doi:10.1111/j.1467-9531.2010.01223.x>

Gile, Krista J., Beaudry, Isabelle S. and Handcock, Mark S., 2018 Methods for Inference from Respondent-Driven Sampling Data, Annual Review of Statistics and Its Application <doi:10.1146/annurev-statistics-031017-100704>.

Gile, Krista J., Handcock, Mark S., 2015 Network Model-Assisted Inference from Respondent-Driven Sampling Data, Journal of the Royal Statistical Society, A. <doi:10.1111/rssa.12091>.

Salganik, M., Heckathorn, D. D., 2004. Sampling and estimation in hidden populations using respondent-driven sampling. Sociological Methodology 34, 193-239.

Volz, E., Heckathorn, D., 2008. Probability based estimation theory for Respondent Driven Sampling. The Journal of Official Statistics 24 (1), 79-97.

See Also

RDS.I.estimates, RDS.II.estimates

Examples

data(fauxmadrona)
RDS.SS.estimates(rds.data=fauxmadrona,outcome.variable="disease",N=1000)

Description

Create RDS samples with given characteristics

Usage

rdssampleC(
  net,
  nnodes = network.size(net),
  nsamp0,
  fixinitial,
  nsamp,
  replace,
  coupons,
  select = NULL,
  bias = NULL,
  rds.samp = NULL,
  seed.distribution = NULL,
  attrall = FALSE,
  trait.variable = "disease",
  nsims = 1,
  seeds = NULL,
  prob.network.recall = 1,
  verbose = TRUE
)

Arguments

Value

A list with the following elements: nsample: vector of indices of sampled nodes wsample: vector of waves of each sampled node degsample: vector of degrees of sampled nodes attrsample: vector of attrs of sampled nodes toattr: vector of numbers of referrals to attrsd nodes tonoattr: vector of number of referrans to unattrsd nominators: recruiter of each sample

Description

This function imports RDSAT data files as rds.data.frame objects.

Usage

read.rdsat(file, delim = c("<auto>", "\t", " ", ","), N = NULL)

Arguments

Examples

fn <- paste0(path.package("RDS"),"/extdata/nyjazz.rdsat")
rd <- read.rdsat(fn)
plot(rd)

Description

Import data saved using write.rdsobj

Usage

read.rdsobj(file)

Arguments

Description

Plots the recruitment network using the Reingold Tilford algorithm.

Usage

reingold.tilford.plot(
  x,
  vertex.color = NULL,
  vertex.color.scale = hue_pal(),
  vertex.size = 2,
  vertex.size.range = c(1, 5),
  edge.arrow.size = 0,
  vertex.label.cex = 0.2,
  vertex.frame.color = NA,
  vertex.label = get.id(x),
  show.legend = TRUE,
  plot = TRUE,
  ...
)

Arguments

Value

A two-column vector of the positions of the nodes in the recruitment tree.

Examples

## Not run: 
data(fauxmadrona)
data(faux)
reingold.tilford.plot(faux)
reingold.tilford.plot(fauxmadrona,vertex.color="disease")

## End(Not run)

Description

Determines the recruiter.id from recruitment coupon information

Usage

rid.from.coupons(
  data,
  subject.coupon = NULL,
  coupon.variables,
  subject.id = NULL,
  seed.id = "seed"
)

Arguments

Examples

fpath <- system.file("extdata", "nyjazz.csv", package="RDS")
dat <- read.csv(fpath)
dat$recruiter.id <- rid.from.coupons(dat,"own.coupon",
                      paste0("coupon.",1:7),"id")

#create and rds.data.frame
rds <- as.rds.data.frame(dat,network.size="network.size")

Description

This function sets the class of the control list, with the default being the name of the calling function.

Usage

set.control.class(
  myname = as.character(RDS::ult(sys.calls(), 2)[[1L]]),
  control = get("control", pos = parent.frame())
)

Arguments

Value

The control list with class set.

See Also

check.control.class, print.control.list

Description

Displays an rds.data.frame

Usage

show.rds.data.frame(x, ...)

Arguments

Description

RDS::summary.svyglm.RDS is a version of summary.svyglm that reports odds-ratios in place of coefficients in the summary table. This only applies for the binomial family. Otherwise it is identical to summary.svyglm. The default in summary.svyglm is to display the log-odds-ratios and this displays the exponetiated from and a 95 p-values are still displayed.

Usage

## S3 method for class 'svyglm.RDS'
summary(object, correlation = FALSE, df.resid = NULL, odds = TRUE, ...)

Arguments

Details

svyglm fits a generalised linear model to data from a complex survey design, with inverse-probability weighting and design-based standard errors.

There is no anova method for svyglm as the models are not fitted by maximum likelihood.

See the manual page on svyglm for detail of that function.

Value

RDS::summary.svyglm returns an object of class "summary.svyglm.RDS", a list with components

See Also

svyglm, summary.

Examples

## For examples see example(svyglm)

Description

calculates the mle. i.e. the row proportions of the transition matrix

Usage

transition.counts.to.Markov.mle(transition.counts)

Arguments

Details

depreicated. just use prop.table(transition.counts,1)

Description

Extract or replace the *ult*imate (last) element of a vector or a list, or an element counting from the end.

Usage

ult(x, i = 1L)

Arguments

Value

An element of 'x'.

Examples

x <- 1:5
(last <- ult(x))
(penultimate <- ult(x, 2)) # 2nd last.

Description

Volz-Heckathorn (RDS-II) weights

Usage

vh.weights(degs, N = NULL)

Arguments

Description

writes an rds.data.frame recruitment tree as a GraphViz file

Usage

write.graphviz(x, file)

Arguments

Description

Writes out the RDS tree in NetDraw format

Usage

write.netdraw(x, file = NULL, by.seed = FALSE)

Arguments

Details

If by.seed is false, two files are created using 'file' as a base name. paste0(file,".DL") contains the edge information, and paste0(file,".vna") contains the nodal attributes

Description

Writes out the RDS tree in RDSAT format

Usage

write.rdsat(x, file = NULL)

Arguments

Description

Export an rds.data.frame to file

Usage

write.rdsobj(x, file)

Arguments