sjstats (version 0.6.0)

boot_ci: Standard Error and Confidence Intervals for bootstrapped estimates

Description

Compute nonparametric bootstrap standard error, confidence intervals and p-value for a vector of bootstrap replicate estimates.

Usage

boot_ci(data, ...)
boot_se(data, ...)
boot_p(data, ...)

Arguments

data

A data frame that containts the vector with bootstrapped estimates, or directly the vector (see 'Examples').

...

Optional, names of the variables with bootstrapped estimates. Required, if either data is a data frame and no vector, or if only selected variables from data should be used in the function.

Value

A data_frame with either bootstrap standard error, the lower and upper confidence intervals or the p-value for all bootstrapped estimates.

Details

This method requires one or more vectors of bootstrap replicate estimates as input. The function then computes the nonparametric bootstrap standard error by calculating the standard deviation of the input vector. The mean value of the input vector and its standard error is used to calculate the lower and upper confidence interval, assuming a t-distribution of bootstrap estimate replicates. P-values from boot_p are also based on t-statistics, assuming normal distribution.

See Also

bootstrap to generate nonparametric bootstrap samples.

Examples

data(efc)
bs <- bootstrap(efc, 100)

# now run models for each bootstrapped sample
bs$models <- lapply(bs$strap, function(x) lm(neg_c_7 ~ e42dep + c161sex, data = x))

# extract coefficient "dependency" and "gender" from each model
bs$dependency <- unlist(lapply(bs$models, function(x) coef(x)[2]))
bs$gender <- unlist(lapply(bs$models, function(x) coef(x)[3]))

# get bootstrapped confidence intervals
boot_ci(bs$dependency)

# compare with model fit
fit <- lm(neg_c_7 ~ e42dep + c161sex, data = efc)
confint(fit)[2, ]

# alternative function calls.
boot_ci(bs$dependency)
boot_ci(bs, dependency)
boot_ci(bs, dependency, gender)


# compare coefficients
mean(bs$dependency)
coef(fit)[2]

# bootstrap() and boot_ci() work fine within pipe-chains
library(dplyr)
efc %>%
  bootstrap(100) %>%
  mutate(models = lapply(.$strap, function(x) {
    lm(neg_c_7 ~ e42dep + c161sex, data = x)
  })) %>%
  mutate(dependency = unlist(lapply(.$models, function(x) coef(x)[2]))) %>%
  boot_ci(dependency)


# check p-value
boot_p(bs$gender)
summary(fit)$coefficients[3, ]


# 'spread_coef()' from the 'sjmisc'-package makes it easy to generate
# bootstrapped statistics like confidence intervals or p-values
library(dplyr)
library(sjmisc)
efc %>%
  # generate bootstrap replicates
  bootstrap(100) %>%
  # apply lm to all bootstrapped data sets
  mutate(models = lapply(.$strap, function(x) {
    lm(neg_c_7 ~ e42dep + c161sex + c172code, data = x)
  })) %>%
  # spread model coefficient for all 100 models
  spread_coef(models) %>%
  # compute the CI for all bootstrapped model coefficients
  boot_ci(e42dep, c161sex, c172code)

# or...
efc %>%
  # generate bootstrap replicates
  bootstrap(100) %>%
  # apply lm to all bootstrapped data sets
  mutate(models = lapply(strap, function(x) {
    lm(neg_c_7 ~ e42dep + c161sex + c172code, data = x)
  })) %>%
  # spread model coefficient for all 100 models
  spread_coef(models, append = FALSE) %>%
  # compute the CI for all bootstrapped model coefficients
  boot_ci()