Data.MathematicalModelingEmployment.xlsx

Unemployment is one of the numerous socio-economic challenges that exist in all countries of the world. Accurate forecasting of the unemployment rate informs significant government and various stakeholders' decisions relating to fiscal and monetary policies, and investment, among others. The main objective of this study was to compare and examine the predictive accuracy of least square, Time series autoregressive distributed lag (ARDL) and Vector Autoregression (VAR) models for predicting the youth unemployment rate of Jordan. Using time series annual data from 1991 to 2021, the three models were fitted to model the youth unemployment rate of Jordan based on Population growth Rate (annual %), Current GDP in US$, Foreign direct investment, net (BoP, current US$), External debt stocks (% of GNI), Foreign direct investment, net inflows (% of GDP), Foreign direct investment, net outflows (% of GDP), Literacy rate, adult total (% of people ages 15 and above), School enrolment, tertiary (gross), gender parity index (GPI), and Inflation, consumer prices (annual %). The predictive accuracy of the models was assessed using mean absolute error, root mean square error and mean absolute percentage error.失业是全球各国普遍存在的诸多社会经济挑战之一。对失业率进行精准预测，可为政府及各类利益相关方制定财政政策、货币政策与投资决策等提供重要依据。 本研究的核心目标为对比并检验最小二乘（Least Square）、时间序列自回归分布滞后模型（ARDL）以及向量自回归模型（VAR）在预测约旦青年失业率时的预测精度。本研究采用1991年至2021年的年度时间序列数据，基于人口增长率（年度百分比）、以美元计价的当期GDP、外国直接投资净额（国际收支平衡表，当期美元）、外债存量（占国民总收入GNI的百分比）、外国直接投资净流入（占GDP的百分比）、外国直接投资净流出（占GDP的百分比）、成人识字率（15岁及以上人口占比）、高等教育毛入学率性别平等指数（GPI）以及消费者物价通胀率（年度百分比）等指标，对上述三种模型进行拟合，以构建约旦青年失业率的预测模型。本研究采用平均绝对误差、均方根误差以及平均绝对百分比误差，对各模型的预测精度进行评估。