latentIndex {hopit}

Calculate the latent index

Description

Calculate the latent index from the fitted model. The latent index is a standardized latent measure that takes values from 0 to 1, where 0 refers to the worst predicted state (the maximal observed value for the latent measure) and 1 refers to the best predicted state (the minimal observed value for the latent measure).

Usage

latentIndex(model, subset = NULL)

healthIndex(model, subset = NULL)

Arguments

model

a fitted hopit model.

subset

an optional vector that specifies a subset of observations.

Value

a vector with a latent index for each individual.

Author(s)

Maciej J. Danko

References

Jurges H (2007). “True health vs response styles: exploring cross-country differences in self-reported health.” Health Economics, 16(2), 163-178. doi:10.1002/hec.1134.

Oksuzyan A, Danko MJ, Caputo J, Jasilionis D, Shkolnikov VM (2019). “Is the story about sensitive women and stoical men true? Gender differences in health after adjustment for reporting behavior.” Social Science & Medicine, 228, 41-50. doi:10.1016/j.socscimed.2019.03.002.

See Also

standardizeCoef, getCutPoints, getLevels, hopit.

Examples

# DATA
data(healthsurvey)

# the order of response levels decreases from the best health to
# the worst health; hence the hopit() parameter decreasing.levels
# is set to TRUE
levels(healthsurvey$health)

# Example 1 ---------------------

# fit a model
model1 <- hopit(latent.formula = health ~ hypertension + high_cholesterol +
                  heart_attack_or_stroke + poor_mobility + very_poor_grip +
                  depression + respiratory_problems +
                  IADL_problems + obese + diabetes + other_diseases,
                thresh.formula = ~ sex + ageclass + country,
                decreasing.levels = TRUE,
                control = list(trace = FALSE),
                data = healthsurvey)

# calculate the health index
hi <- latentIndex(model1)

summary(hi)

# plot a simple histogram of the function output
hist(hi, col='deepskyblue3')

#plot the reported health status versus the health index.
plot(hi, response = "data", ylab = 'Health index',
     col='deepskyblue3', main = 'Reported health levels')

# plot the model-predicted health levels versus the health index.
plot(hi, response = "fitted", ylab = 'Health index',
     col='deepskyblue3', main = 'Model-predicted health levels')