: Linear Cointegration

Showing results in 'Publications'. Show all posts
Shem Otoi Sam, Manene MM, Isaac C Kipchirchir, Pokhariyal GP. "Cointegration analysis of youth unemployment in Kenya." International Journal of Statistics and Applied Mathematics. 2020;5(3): 129-133. AbstractWebsite

In this paper analysis of contribution of macroeconomic variables gross domestic product
(GDP), external debt (ED), foreign direct investment (FDI), private investment (PI), youth
population (POP), and youth literacy rate (LR) to youth unemployment (YUN) in Kenya over
time is done. The analysis is done under framework of cointegration of time series data. First,
logarithmic transformation of the series is carried out followed by stationarity test to determine
the order of stationarity. The Philip-Ouliaris cointegration test is carried out to determine
whether the series are individually cointegrated in a pair-wise manner. Then the Johansen
cointegration test is conducted to determine the rank of cointegration. The paper does not
proceed to identify cointegration relations as that is superfluous as far as estimation of linear
cointegration model is concerned. Finally the linear cointegration equation of the
macroeconomic variables is estimated and interpreted. Philip-Ouliaris test reveals that six pairs
are I(0) while 15 pairs are I(1). The Augmented Dickey-Fuller test finds that GDP, FDI, and
ED are stationary at level, i.e. without differencing whereas PI, LR, YUN, and POP are
stationary of first difference. According to Johansen cointegration test, the rank of
cointegration is 3, revealing three cointegration relations among the variables used. The results
indicate that 1% increase in GDP, ED, FDI, and LR increases YUN by 0.356204%, 0.269%,
0.002441%, and 0.154216 respectively. Contrarily, 1% increase in population reduces youth
unemployment by 0.350833%.The model is subjected to F-test and p-value test and found to
be statistically significant.

UoN Websites Search