Adeboye, Nurain Olawale and Alabi, Nurudeen Olawale (2022) Deep-LearningModellingofDynamicPanelDataforAfricanEconomicGrowth. Journal of Econometrics and Statistics, 2 (1). pp. 47-60. ISSN 2583-0473
![[img]](https://eprints.federalpolyilaro.edu.ng/style/images/fileicons/text.png) |
Text
Deep-Learning Modelling of Dynamic Data for African Economic Growth.pdf Download (404kB) |
Abstract
NigeriaApril28,2022Whenmodellingphenomenarelatingtotheeconomy,dynamicpanelmodelshaveproventobeausefultool.Previousstudieshavemodelleddynamicpaneldatausingconventionalmethodsofgeneralizedmethodofmoment,InstrumentalvariablesandMaximumlikelihoodestimatorsamongothers.Thisstudyhoweverfocusesonmodellingdynamicpaneldatausingmoderndayapproachesofdeep-learningtechniques.Tothisend,twomacro-economicvariablesofPurchasingPowerParity(PPP)andGrossNationalIncome(GNI)wereemployedtomodeltheeco-nomicgrowthoftwentyAfricancountries.DynamicpanelinformationaboutthesecountriesweresourcedfromUNESCOdatabasebetween1990and2019.DeeplearningtechniquesofLongTermShortmemory(LSTM),BidirectionalLongShortTermMemory(Bi-LSTM)andGatedRecurrentUnits(GRU)wereemployedinthemodellingprocess,andthefindingsrevealedthatLSTMhavingtheleastvaluesoftheadoptedeval-uationmetrics,isthebestandmostsuitabledeeplearningmethodformodellingdynamicpaneldata.Forecastswerealsomadeforthenext20yearswiththetechniques,andtheresultsshowthatLSTMgivesthebestpredictingaccuracywithitslowestMeanAbsoluteError(MAE),MAPE,MSEandRMSE.
| Item Type: | Article |
|---|---|
| Subjects: | Q Science > QA Mathematics |
| Divisions: | Faculty of Engineering, Science and Mathematics > School of Electronics and Computer Science |
| Depositing User: | Mr. Bolanle Yisau I. |
| Date Deposited: | 31 Aug 2023 04:37 |
| Last Modified: | 31 Aug 2023 04:37 |
| URI: | http://eprints.federalpolyilaro.edu.ng/id/eprint/2330 |
Actions (login required)
 |
View Item |
