These smoothing functions allow smoothing of a variable in a vital object.
The vital object is returned along with some additional columns containing
information about the smoothed variable: usually .smooth containing the
smoothed values, and .smooth_se containing the corresponding standard errors.
Usage
smooth_spline(.data, .var, age_spacing = 1, k = -1)
smooth_mortality(.data, .var, age_spacing = 1, b = 65, power = 0.4, k = 30)
smooth_fertility(.data, .var, age_spacing = 1, lambda = 1e-10)
smooth_loess(.data, .var, age_spacing = 1, span = 0.2)Arguments
- .data
A vital object
- .var
name of variable to smooth
- age_spacing
Spacing between ages for smoothed vital. Default is 1.
- k
Number of knots to use for penalized regression spline estimate.
- b
Lower age for monotonicity. Above this, the smooth curve is assumed to be monotonically increasing.
- power
Power transformation for age variable before smoothing. Default is 0.4 (for mortality data).
- lambda
Penalty for constrained regression spline.
- span
Span for loess smooth.
Details
smooth_mortality() use penalized regression splines applied to log mortality
with a monotonicity constraint above age b. The methodology is based on Wood (1994).
smooth_fertility() uses weighted regression B-splines with a concavity constraint,
based on He and Ng (1999). The function smooth_loess() uses locally quadratic
regression, while smooth_spline() uses penalized regression splines.
References
Hyndman, R.J., and Ullah, S. (2007) Robust forecasting of mortality and fertility rates: a functional data approach. Computational Statistics & Data Analysis, 51, 4942-4956. https://robjhyndman.com/publications/funcfor/
Examples
library(dplyr)
#>
#> Attaching package: ‘dplyr’
#> The following objects are masked from ‘package:stats’:
#>
#> filter, lag
#> The following objects are masked from ‘package:base’:
#>
#> intersect, setdiff, setequal, union
aus_mortality |>
filter(State == "Victoria", Sex == "female", Year > 2000) |>
smooth_mortality(Mortality)
#> # A vital: 2,020 x 10 [1Y]
#> # Key: Age x (Sex, State, Code) [101 x 1]
#> Year Age Sex State Code Mortality Exposure Deaths .smooth .smooth_se
#> <int> <dbl> <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl[1> <dbl[1d]>
#> 1 2001 0 female Victor… VIC 0.00404 29229. 118. 4.02e-3 0.000347
#> 2 2001 1 female Victor… VIC 0.000405 29654. 12 3.90e-4 0.0000789
#> 3 2001 2 female Victor… VIC 0.000201 29832. 6 2.16e-4 0.0000404
#> 4 2001 3 female Victor… VIC 0.000134 29859. 4.01 1.54e-4 0.0000287
#> 5 2001 4 female Victor… VIC 0.000165 30328. 5.01 1.25e-4 0.0000233
#> 6 2001 5 female Victor… VIC 0.0000652 30698. 2 1.10e-4 0.0000205
#> 7 2001 6 female Victor… VIC 0.0000959 31286. 3 1.02e-4 0.0000189
#> 8 2001 7 female Victor… VIC 0.0000945 31748. 3 9.96e-5 0.0000181
#> 9 2001 8 female Victor… VIC 0.0000943 31810. 3 1.01e-4 0.0000178
#> 10 2001 9 female Victor… VIC 0.0000627 31893. 2 1.05e-4 0.0000180
#> # ℹ 2,010 more rows
aus_fertility |>
filter(Year > 2000) |>
smooth_fertility(Fertility)
#> # A vital: 210 x 7 [1Y]
#> # Key: Age [35 x 1]
#> Year Age Fertility Exposure Births .smooth .smooth_se
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 2001 15 0.00320 132027 423. 0.00320 0.00333
#> 2 2001 16 0.00728 133096 969. 0.00755 0.00709
#> 3 2001 17 0.0158 131433 2075. 0.0158 0.0133
#> 4 2001 18 0.0249 133123 3313. 0.0260 0.0196
#> 5 2001 19 0.0372 132398 4931. 0.0357 0.0239
#> 6 2001 20 0.0435 131377 5721. 0.0435 0.0258
#> 7 2001 21 0.0485 127985 6202. 0.0499 0.0261
#> 8 2001 22 0.0581 126901 7373. 0.0572 0.0262
#> 9 2001 23 0.0656 127134 8336. 0.0656 0.0263
#> 10 2001 24 0.0749 128239 9599. 0.0749 0.0264
#> # ℹ 200 more rows