|
Abstract:
The relationship of the unemployment rate (%) and the educational budget per university student is discussed in this paper. The university students in Taiwan enjoy the highest educational budget per person and most of them will join to the job markets after graduation. Based on these factors, the educational budget per student (in 1,000 USD) was chosen to study its relationship with the unemployment rate. The time series data from 1978 to 2015 were obtained from DGBAS public archive. After using augment Dickey-Fuller (ADF) test, the educational budget per university student and unemployment rate are not stationary, but after first difference both are stationary, and can be denoted as I(1). After regression, a positive parameter 0.4598 was obtained. It means increasing one unit of educational budget will increase 0.4598 unit of unemployment rate. A further check into the residuals of the regression by the cointegration test shows the residuals are not stationary (not cointegrated). In other words, the regression equation between unemployment rate and educational budget per university student is spurious. The more complicated vector autoregressive (VAR) model then was used to solve this spurious problem. The VAR model, which also got the change of unemployment rate at time t, has positive relationship with the change of the educational budget at time t-1. However, all the parameters obtained by the VAR model are insignificant in 5% level. In other words, even the regression equations obtained by the VAR model are unreliable either. The only explanation is that the unemployment rate and educational budget per university student has very slim relationship. It seems that the Ministry of Education (MOE) in Taiwan seems should remove the requirement of "employment rate" of a school's graduated students as an indicator in its subsidy program. Because there is no or just a slim relationship between unemployment rate and educational budget.
|
