Page 72 - Azerbaijan State University of Economics

THE JOURNAL OF ECONOMIC SCIENCES: THEORY AND PRACTICE, V.83, # 1, 2026, pp. 58-81

Porter, 2009; Wooldridge, 2010). The only variables lacking stationarity are TEN,
internet, and EST_BUS, while the remaining variables are found to be stationary
natural logs. The first differentiation renders these variables stationary at I (1)
according to the results, as depicted in Table 3.

This study then verifies that TEN and ICT networks constitute a single integrated pattern
to determine whether there is a long-run equilibrium relationship between entrepreneurial
activity and digital infrastructure. Co-integration is a crucial concept in panel data
analysis, as it provides confidence that, although variables may not be stationary on their
own, they may still move together in a steady, long-term relationship (Westerlund, 2007;
Engle & Granger, 1987). The research assessed integration between TEN, ICT, Internet,
RR&D, R&D, EST_BUS, GDP, and Financial Risk through Westerlund's approach. This
research applied Westerlund's co-integration procedure (2007) to examine both cross-
sectional relationships and distinct patterns in the available data. Interdependence and
variations across the data. The findings from the Westerlund. A co-integration report in
Table 4 shows the results.

Table 4: Co-integration analysis
Model Stat Value Z stat

Model 1 GT -4.797 -14.999
JA -13.888 -1.777
PT -23.999 -12.099
PA 13.887 -2.998
Model 2 GT -4.534 -13.676
Ga -14.234 -2.565
PT -23.001 -13.078
PA 14.002 -1.778
Author’s Creation

Table 4 shows the results of the co-integration test of Westerlund (2007) based on GT
(Group-mean test statistic), GA (Group-mean ADF statistic), PT(Panel statistic), and
PA(Panel ADF statistic) statistics. Both models have significant test value (p < 0.01),
thus establishing that there is a long-run equilibrium relationship between technology
entrepreneurship (TEN) and its explanatory variables. This confirms the suitability of
these variables for use in long-run panel regressions.
A causality analysis on Table 5 shows how ICT exports relate to economic activities.
To obtain insight into the direction of causality between the main explanatory
variables and digital entrepreneurship (TEN), Granger causality analysis was
conducted. As indicated in the test results (Table 5), the chosen variables have a
statistically significant causal relationship with TEN at the 1 percent level (p < 0.01).

72