Package 'WaveletKNN'

Title: Wavelet Based K-Nearest Neighbor Model
Description: The employment of the Wavelet decomposition technique proves to be highly advantageous in the modelling of noisy time series data. Wavelet decomposition technique using the "haar" algorithm has been incorporated to formulate a hybrid Wavelet KNN (K-Nearest Neighbour) model for time series forecasting, as proposed by Anjoy and Paul (2017) <DOI:10.1007/s00521-017-3289-9>.
Authors: Dr. Ranjit Kumar Paul [aut], Dr. Md Yeasin [aut, cre]
Maintainer: Dr. Md Yeasin <[email protected]>
License: GPL-3
Version: 0.1.0
Built: 2025-02-24 04:03:27 UTC
Source: https://github.com/cran/WaveletKNN

Help Index

Wavelet Based K-Nearest Neighbor Model

Description

Wavelet Based K-Nearest Neighbor Model

Usage

WaveletKNN(ts, MLag = 12, split_ratio = 0.8, wlevels = 3)

Arguments

ts

Time Series Data
MLag

Maximum Lags
split_ratio

Training and Testing Split
wlevels

Number of Wavelet Levels

Value

  • Lag: Lags used in model

  • Parameters: Parameters of the model

  • Train_actual: Actual train series

  • Test_actual: Actual test series

  • Train_fitted: Fitted train series

  • Test_predicted: Predicted test series

  • Accuracy: RMSE and MAPE of the model

References

  • Aminghafari, M. and Poggi, J.M. 2012. Nonstationary time series forecasting using wavelets and kernel smoothing. Communications in Statistics-Theory and Methods, 41(3),485-499.

  • Paul, R.K. A and Anjoy, P. 2018. Modeling fractionally integrated maximum temperature series in India in presence of structural break. Theory and Applied Climatology 134, 241–249.

Examples

library("WaveletKNN")
data<- rnorm(100,100, 10)
WG<-WaveletKNN(ts=data)