Short-term forecasts of Economic Growth for Nigeria using the Average Centered Penalized Regression Method

Author:

Ebiwonjumi Ayooluwade1,Chifurira Retius2,Chinhamu Knowledge2

Affiliation:

1. Adekunle Ajasin University Akungba Akoko

2. University of kwaZulu-Natal

Abstract

Abstract

in this study, we predicted ten (10) quarterly economic growth rate represented by RGDP the presence of multicollinearity and outliers. We employed the average centered penalized regression techniques comprising of LASSO, ridge and elastic net method to quarterly data from 1986 to 2021 obtained from the Central Bank of Nigeria. Consequently, the predictive power for the fitted LASSO, ridge and elastic net penalized least square model using RMSE were revealed to be 0.2895, 0.3197 and 0.2979 respectively, MAE were revealed to be 0.2174, 0.2400 and 0.2237 respectively, MAPE were revealed to be 2.1298, 2.3489 and 2.1985 respectively. The R Square values indicated that economic growth drivers identified can explained 91.9%, 91.0% and 92.0% variation or changes in the economic growth in Nigeria when LASSO, ridge and elastic net as penalized least square regression techniques were respectively adopted. The lambda penalty for the fitted model using LASSO technique was 0.084 which was higher than that of elastic net technique that stood at 0.000. Also, based on smallest RMSE, MAE and MAPE, it can be emphasized that LASSO as an average centered penalized regression techniques was more sensitive and efficient for estimating and predicting the economic growth when multicollinearity and outliers were jointly present in data set. Hence, the LASSO method showed that INDT, RINR, REXR and OPEN were positively contributed to the RGDP to the turn of 4.27%. 0.40%, 0.49% and 0.53% respectively and the contribution of EXDT to economic growth was negative and as such led to decline in RGDP by 0.98% during the period under consideration. Thus, it can be concluded based on RMSE, MAE and MAPE, normal quantile plot and quantile predicted plot as well as the estimated parameter of RGDP that LASSO was the best and most appropriate model for predicting economic situation in Nigeria. This was evidently showed by the stability of the series and forecasted 10 quarterly RGDP. Therefore, there is a need for urgent action and strong willed to address the retrogressive economic situation. Thus, economic openness should be adopted to enhance the development and the growth of economy.

Publisher

Springer Science and Business Media LLC

Reference35 articles.

1. Trade policy and economic growth in Nigeria;Afolabi AI;Economic Thought Journal,2020

2. The relationship between exports and economic growth in Saudi Arabia;Alkhateeb TTY;Asian Social Science,2016

3. External debt and economic growth: The case of emerging economy;Al Kharusi S;Journal of Economic Integration,2018

4. Monetary policy and economic growth in Nigeria: an ARDL-bound testing and ECM approach;Ayomitunde AT;European Journal of Economics Law Politics,2018

5. Econometric analysis of external debt, exchange rate and economic growth in Nigeria;Ayunku PE;Researchers World – Journal of Arts, Science & Commerce,2016

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3