Forecasting Domestic Tax Revenues in Kenya Using Sarima & Holt-Winters Methods
Keywords:
Kenya Revenue Authority(KRA), Domestic Taxes Revenues, Holt-Winters (HW), Seasonal Autoregressive Integrated Moving Average (SARIMA)Abstract
Forecasting of tax revenues is an important factor in fiscal planning. Underestimation and overestimation of tax revenues lead to unstable economies. The study sought to find suitable Holt-Winters and SARIMA models that could be used to forecast Domestic tax revenues in Kenya. The study utilized the Domestic tax revenues collected in Kenya between Jan 2015 to December 2020. Analysis of data was done using R-software (version 4.1.0) where SARIMA and Holt-Winters time series forecasting methods were applied to the revenue data. SARIMA(0,1,1)(0,1,1)[12] model was found to be the best model since it had the least Bayesian Information Criterion (BIC=1236.49) and least forecasting errors (MAPE=6.9, MASE=0.37).The multiplicative Holt-Winters method was slightly superior to the additive method due to its lower error (MAPE=7.43). The study recommends the use of the two methods to forecast Domestic taxes in Kenya be used to capture the Domestic taxes revenues with high precision.
Downloads
Published
Issue
Section
License
Copyright (c) 2022 Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.