Type: | Package |
Title: | Support Functions for Time Series Analysis Book |
Version: | 0.1.0 |
Description: | Contains the support functions for the Time Series Analysis book. We present a function to calculate MSE and MAE for inputs of actual and forecast values. We also have the code for disaggregation as found in Wei and Stram (1990, <doi:10.1111/j.2517-6161.1990.tb01799.x>), and Hodgess and Wei (1996, "Temporal Disaggregation of Time Series"). |
Depends: | R (≥ 4.4.0), PolynomF |
License: | GPL-2 | GPL-3 |
Encoding: | UTF-8 |
RoxygenNote: | 7.3.1 |
NeedsCompilation: | no |
Packaged: | 2025-02-22 07:01:14 UTC; e |
Author: | Erin Hodgess [aut, cre] |
Maintainer: | Erin Hodgess <erinm.hodgess@gmail.com> |
Repository: | CRAN |
Date/Publication: | 2025-02-24 17:20:05 UTC |
Create a disaggregated time series
Description
Input an annual, quarterly series. Create a quarterly or monthly series via ARIMA
Usage
disag1(x, m)
Arguments
x |
Input ts, must have frequency of 1 or 4 |
m |
Order of disaggregation, must be 12, 4, or 3 |
Details
Uses ARIMA model on the aggregate series to create a disaggregate series
Value
y_s |
Disag. series to be summed |
y_m |
Disag. series mean |
disphi |
Disagg phi value |
distheta |
Disagg theta value |
dissig2 |
Disagg sigma2 |
References
William W.S. Wei and Daniel Stram, 1990, Disaggregation of Time Series Models, Journal of the Royal Statistics Society, B, Vol 52, Number 3, pp. 453-467. Erin M. Hodgess and William W.S. Wei, 1996, Temporal Disaggregation of Time Series, Applied Statistical Science I, pp. 33-43, Nova Science Publishers, Commack, NY
Calculate MSE and MAE for actual and forecast values
Description
The inputs are the actual and the forecast values. We calculate the Mean Square Error (MSE) and Mean Absolute Error (MAE)
Usage
foremeas1(actx, forex)
Arguments
actx |
actual values |
forex |
forecast values |
Details
MSE = mean((act-fore)^2), MAE = mean(abs(act-fore))
Value
MSE |
Mean square error |
MAE |
Mean absolute error |
Author(s)
c( person( "Erin", "Hodgess", email = "erinm.hodgess@gmail.com", role = c("aut", "cre") ) )
Create an nxn symmetric matrix from an n length vector
Description
Create an nxn symmetric matrix from an n length vector
Usage
mySym(x)
Arguments
x |
input length n vector |
Details
create an nxn symmetric matrix
Value
y |
symmetric matrix |
Author(s)
c( person( "Erin", "Hodgess", email = "erinm.hodgess@gmail.com", role = c("aut", "cre") ) )
Examples
##---- Should be DIRECTLY executable !! ----
##-- ==> Define data, use random,
##-- or standard data sets, see data().
mySym(1:6)