Title: | Two Stage Forecasting (TSF) for Long Memory Time Series in Presence of Structural Break |
---|---|
Description: | Forecasting of long memory time series in presence of structural break by using TSF algorithm by Papailias and Dias (2015) <doi:10.1016/j.ijforecast.2015.01.006>. |
Authors: | Sandipan Samanta, Ranjit Kumar Paul and Dipankar Mitra |
Maintainer: | Dr. Ranjit Kumar Paul <[email protected]> |
License: | GPL |
Version: | 0.1.1 |
Built: | 2025-03-12 03:30:45 UTC |
Source: | https://github.com/cran/TSF |
The function is used for forecasting long memory time series using TSF approach
forecastTSF(N0,Xt,bandwidth)
forecastTSF(N0,Xt,bandwidth)
N0 |
lead period of forecast |
Xt |
univariate time series |
bandwidth |
the bandwidth used in the regression equation |
forecastTSF |
the predicted values, the out of sample forecasts and the values of long memory parameter |
Sandipan Samanta, Ranjit Kumar Paul and Dipankar Mitra
Papailias, F. and Dias, G. F. 2015. Forecasting long memory series subject to structural change: A two-stage approach. International Journal of Forecasting, 31, 1056 to 1066.
Wang, C. S. H., Bauwens, L. and Hsiao, C. 2013. Forecasting a long memory process subject to structural breaks. Journal of Econometrics, 177, 171-184.
Reisen, V. A. (1994) Estimation of the fractional difference parameter in the ARFIMA(p,d,q) model using the smoothed periodogram. Journal Time Series Analysis, 15(1), 335 to 350.
## Simulating Long Memory Series N <- 1000 PHI <- 0.2 THETA <- 0.1 SD <- 1 M <- 0 D <- 0.2 Seed <- 123 N0<-9 bandwidth<-0.9 set.seed(Seed) Sim.Series <- fracdiff::fracdiff.sim(n = N, ar = c(PHI), ma = c(THETA), d = D, rand.gen = rnorm, sd = SD, mu = M) Xt <- as.ts(Sim.Series$series) ## Forecasting using TSF method forecastTSF (N0,Xt,bandwidth)
## Simulating Long Memory Series N <- 1000 PHI <- 0.2 THETA <- 0.1 SD <- 1 M <- 0 D <- 0.2 Seed <- 123 N0<-9 bandwidth<-0.9 set.seed(Seed) Sim.Series <- fracdiff::fracdiff.sim(n = N, ar = c(PHI), ma = c(THETA), d = D, rand.gen = rnorm, sd = SD, mu = M) Xt <- as.ts(Sim.Series$series) ## Forecasting using TSF method forecastTSF (N0,Xt,bandwidth)
The function is used for prediction of long memory time series in presence of structural break
StructuralBrekwithLongmemory(ts,bandwidth)
StructuralBrekwithLongmemory(ts,bandwidth)
ts |
univariate time series |
bandwidth |
the bandwidth used in the regression equation |
StructuralBrekwithLongmemory |
the updated series at first step of TSF appraoch, prediction based on TSF approach and the estimate of long memory parameter |
Sandipan Samanta, Ranjit Kumar Paul and Dipankar Mitra
Papailias, F. and Dias, G. F. 2015. Forecasting long memory series subject to structural change: A two-stage approach. International Journal of Forecasting, 31, 1056 to 1066.
Wang, C. S. H., Bauwens, L. and Hsiao, C. 2013. Forecasting a long memory process subject to structural breaks. Journal of Econometrics, 177, 171-184.
Reisen, V. A. (1994) Estimation of the fractional difference parameter in the ARFIMA(p,d,q) model using the smoothed periodogram. Journal Time Series Analysis, 15(1), 335 to 350.
## Simulating Long Memory Series N <- 1000 PHI <- 0.2 THETA <- 0.1 SD <- 1 M <- 0 D <- 0.2 Seed <- 123 bandwidth<-0.9 set.seed(Seed) Sim.Series <- fracdiff::fracdiff.sim(n = N, ar = c(PHI), ma = c(THETA), d = D, rand.gen = rnorm, sd = SD, mu = M) Xt <- as.ts(Sim.Series$series) ## Forecasting using TSF method StructuralBrekwithLongmemory(Xt,bandwidth)
## Simulating Long Memory Series N <- 1000 PHI <- 0.2 THETA <- 0.1 SD <- 1 M <- 0 D <- 0.2 Seed <- 123 bandwidth<-0.9 set.seed(Seed) Sim.Series <- fracdiff::fracdiff.sim(n = N, ar = c(PHI), ma = c(THETA), d = D, rand.gen = rnorm, sd = SD, mu = M) Xt <- as.ts(Sim.Series$series) ## Forecasting using TSF method StructuralBrekwithLongmemory(Xt,bandwidth)
The function fdseries computes the fractional differenced series for any value of d i.e. positive or negetive.
fdseries(x, d)
fdseries(x, d)
x |
univariate time series |
d |
The orer of fractional differencing to be done |
fdseries |
fractionally differenced series for both positive as well as negetive d |
Sandipan Samanta, Ranjit Kumar Paul and Dipankar Mitra
Papailias, F. and Dias, G. F. 2015. Forecasting long memory series subject to structural change: A two-stage approach. International Journal of Forecasting, 31, 1056 to 1066.
Wang, C. S. H., Bauwens, L. and Hsiao, C. 2013. Forecasting a long memory process subject to structural breaks. Journal of Econometrics, 177, 171-184.
Reisen, V. A. (1994) Estimation of the fractional difference parameter in the ARFIMA(p,d,q) model using the smoothed periodogram. Journal Time Series Analysis, 15(1), 335 to 350.
## Simulating Long Memory Series N <- 1000 PHI <- 0.2 THETA <- 0.1 SD <- 1 M <- 0 D <- 0.2 Seed <- 123 set.seed(Seed) Sim.Series <- fracdiff::fracdiff.sim(n = N, ar = c(PHI), ma = c(THETA), d = D, rand.gen = rnorm, sd = SD, mu = M) Xt <- as.ts(Sim.Series$series) ## fractional differencing fdseries(Xt,d=D)
## Simulating Long Memory Series N <- 1000 PHI <- 0.2 THETA <- 0.1 SD <- 1 M <- 0 D <- 0.2 Seed <- 123 set.seed(Seed) Sim.Series <- fracdiff::fracdiff.sim(n = N, ar = c(PHI), ma = c(THETA), d = D, rand.gen = rnorm, sd = SD, mu = M) Xt <- as.ts(Sim.Series$series) ## fractional differencing fdseries(Xt,d=D)