Skip to contents

Methods to retrieve and display the results of a model estimated with fitPrevalenceModel.

Usage

# S3 method for class 'prevalence_model'
coef(object, ...)

# S3 method for class 'prevalence_model'
fitted(object, ...)

# S3 method for class 'prevalence_model'
residuals(object, ...)

# S3 method for class 'prevalence_model'
as.data.frame(
  x,
  row.names = NULL,
  optional = FALSE,
  what = c("coefficients", "fitted", "observed", "fit_stats", "lag_search", "shifts"),
  ...
)

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

# S3 method for class 'prevalence_model'
summary(object, ...)

# S3 method for class 'summary.prevalence_model'
print(x, ...)

Arguments

object, x

a model estimated with fitPrevalenceModel

...

not used

row.names, optional

not used

what

table to be returned by as.data.frame() (see above)

Value

a data frame, a vector, or (for summary) a list of class "summary.prevalence_model"

Details

  • coef() returns the coefficients in a data frame (long format), one row per series, age and term: "intercept" (value of the transformed prevalence at the last year of the estimation) and "slope" (annual change) for method "trend"; "alpha" (age effects) and "beta" (slope on the mortality rates, with age = NA if it is common to all ages) for method "mortality". For method "shift", "shift" (speed of the horizontal shift of the curve of prevalences, in years of age per calendar year, with age = NA). For method "target", "reference" (value of the indicator at the reference year) for rules "constant" and "horizon", and "intercept" and "slope" (value and annual change of the indicator) for rule "trend"; with scope "min_age", only the minimum age (or 'anchor_age') is concerned. Columns: variables defining series, 'age', 'term', 'estimate', 'std.error', 'n'.

  • fitted() and residuals() return fitted prevalences and residuals (on the scale of the estimation) as vectors, in the order of the rows of as.data.frame(x, what = "fitted").

  • as.data.frame() returns one of the tables of the model, selected by what: "coefficients", "fitted" (observed and fitted prevalences and residuals by year and age), "observed" (data used for the estimation), "fit_stats" (number of observations, residual standard deviation and R-squared for each series) or "lag_search" (sum of squared residuals for each value of 'age_lag' tried, for method "mortality" when 'age_lag' is a vector) or "shifts" (for method "shift": shift of the curve of prevalences estimated for each year, 'shift_obs', and shift predicted by the model, 'shift_fit'). These tables can be used directly to draw graphics.

  • summary() and print() display a short description of the model.

Examples

ages <- 60:100; years <- 2010:2019
obs <- expand.grid(age = ages, year = years)
obs$pix <- plogis(-6 + 0.12 * (obs$age - 60) - (0.01 + 0.0005 * (obs$age - 60)) * (obs$year - 2019))
m <- fitPrevalenceModel(obs, method = "trend")
summary(m)
#> Model of prevalences by age, method "trend" (scale: logit)
#> Years used for the estimation: 2010 to 2019  
#> 
#> Fit statistics by series (sigma: residual standard deviation on the scale of the estimation):
#>  n_obs n_ages n_years     sigma r2
#>    410     41      10 3.291e-16  1
head(coef(m))
#>   age      term estimate    std.error  n
#> 1  60 intercept    -6.00 2.610151e-16 10
#> 2  61 intercept    -5.88 0.000000e+00 10
#> 3  62 intercept    -5.76 0.000000e+00 10
#> 4  63 intercept    -5.64 0.000000e+00 10
#> 5  64 intercept    -5.52 0.000000e+00 10
#> 6  65 intercept    -5.40 5.536967e-16 10
head(as.data.frame(m, what = "fitted"))
#>   year age         pix     pix_fit         resid
#> 1 2010  60 0.002704851 0.002704851  0.000000e+00
#> 2 2010  61 0.003062367 0.003062367  0.000000e+00
#> 3 2010  62 0.003466974 0.003466974  0.000000e+00
#> 4 2010  63 0.003924829 0.003924829  0.000000e+00
#> 5 2010  64 0.004442879 0.004442879  0.000000e+00
#> 6 2010  65 0.005028962 0.005028962 -8.881784e-16