smoo_dataset: Smoother based on weighting

In federico-m-stefanini/convergEU: Monitoring Convergence of EU Countries

Smoother based on weighting

Description

The smoother substitutes an original raw value $y_m,i,t$ of country $m$ indicator $i$ at time $t$ with the weighted average $$\checky_m,i,t = y_m,i,t-1 ~ (1-w)/2 +w ~y_m,i,t +y_m,i,t+1 ~(1-w)/2$$, where $0< w \leq 1$. The special case $w=1$ corresponds to no smoothing. In case of missing values an NA is returned. If the weight is outside the interval $(0,1]$ then a NA is returned. The first and last values are smoothed using weights $w$ and $1-w$.

Usage

smoo_dataset(myTB, leadW = 1, timeTB = NULL)

Arguments

myTB	a complete dataset time by countries, with just country columns.
leadW	leading positive weight less or equal to 1.
timeTB	a dataset with the time variable, if a dataset is desired as output

Value

a matrix of dataset of smoothed values

References

https://local.disia.unifi.it/stefanini/RESEARCH/coneu/tutorial-conv.html

Examples

# Example 1
# Dataset in the format time by countries:
myTB  <- tibble::tibble(
    time = 2001:2010,
    IT = c(10,14,13,12,9,11,13,17,15,25),
    DE = c(10,11,12,9,14,17,23,29,26,23)
    )

# Remove the time variable in order to obtain just country columns and compute smoothed values:
reSMO <- smoo_dataset(myTB[,-1], leadW=1)
reSMO1 <- smoo_dataset(myTB[,-1], leadW=0.5)

# Add the time variable for tibble in output:
reSMO2 <- smoo_dataset(myTB[,-1], leadW=.5,timeTB= dplyr::select(myTB,time))

# Example 2
# Smoother based on weighting for the emp_20_64_MS Eurofound dataset:
data(emp_20_64_MS)
# Select countries:
myTB <- dplyr::select(emp_20_64_MS, time, IT,DE,FR)
# Compute smoothed values by also adding the time variable to the output:
resSM <- smoo_dataset(dplyr::select(myTB,-time), leadW = 0.2, timeTB= dplyr::select(myTB,time))

federico-m-stefanini/convergEU documentation built on July 30, 2023, 3:22 a.m.