Global Journal of Human Social Science, E: Economics, Volume 22 Issue 2

If the data have no unit root [they are said to be level stationary and denoted I(0)], or if they do, a differentiation is all that is required to make them stationary [they are said to be integrated at order one and marked I(1)]. We note that all of the data in our model are I(1), implying that they are high quality. The second prerequisite leads us to use the Johansen (1988) cointegration test, which produces the following result in Eviews 8.1: Table 3: Johansen cointegration test Series: TxINF SYSTINS TxMASSMO TDI Lags interval (in first difference) 1 to 1 Unrestricted Cointegration Rank Test (TRACE) Hypothesized no. Of CE(s) Eigenvalue Trace statistic 0.05 Critical value Prob.** None * 0.856677 61.62443 40.17493 0.0001 At most 1 0.279107 13.05814 24.27596 0.0167 At most 2 0.164332 4.876509 12.32090 0.5849 At most 3 0.015417 0.388424 4.129906 0.0964 Trace test indicates no cointegration at the 0.05 level *denotes rejection of the hypothesis at the 0.05 level **MacKinnon-Haug-Michelis (1999) p-value Source: Authors computation, using Eviews software Only the "none" hypothesis (no association between the variables) is marked with a starin the table above, and thus must be rejected: our variables have a long-term relationship. We can now proceed with our model's regression after confirming the two conditions for estimation. Table 4: Estimation by MCOs Dependent Variable: Log(Tx )Method: Least Squares Date: 03/08/21 Time: 06:14Sample: 1990 2019 Included observations: 30 Variable Coefficient Std.Error t-statistic Prob. C 3, 359792 0,488038 6,884285 0,0000*** Log(SYSTINS) -0,185110 0,232562 -0,795959 0,0370** Log(TxICI) 0,055689 0,085495 0,651377 0,0042*** Log(TxIDE) -0,249589 0,232562 -2,085935 0,0116** R-squared 0,719210 Mean dependent var 0,524214 Adjusted R-squared 0,686811 S.D. dependent var 0,240694 S.E of regression 0,134700 Kaike info criterion -1,047961 SumSquared resid 0,471749 Schwarz criterion -0,861135 Log likelihood 19,71942 Hannan-Quinn -0,988194 F-Statistic 22,19858 Durbin watson 1,393954 Prob(F-statistic) 0,000000 Prob(Wald F-statistic) 0,000000 *, **, *** indicate significance at the 10%, 5% and 1% thresholds respectively Source: Authors computation, using Eviews software © 2022 Global Journals Volume XXII Issue II Version I 35 ( ) Global Journal of Human Social Science - Year 2022 E Institutional Analysis of the Determinants of Economic Non-Take-Off and High Living Standards in Cameroon between 1990 and 2019