Page 120 - AJWEP-v22i3

Nabi, et al.

Table 6. Diagnostic test estimates
Test Statistic p‑value Conclusion
Breusch–Godfrey LM (serial correlation) 1.84 0.17 No serial correlation
Breusch–Pagan–Godfrey (heteroskedasticity) 2.15 0.14 No heteroskedasticity
Jarque–Bera (Normality) 1.73 0.42 Residuals are normally distributed
Ramsey RESET (specification) 1.92 0.16 Model is correctly specified
CUSUM — — Stable within 5% significance bounds
CUSUMSQ — — Stable within 5% significance bounds
Source: Author’s estimate.
Abbreviations: CUSUM: Cumulative sum; CUSUMSQ: Cumulative sum of squares.

Table 7. ARDL bounds estimates attributed to the intensified agricultural techniques
Test statistic Value k rather than the environmental benefits. An increase
F-statistic 10.43256 7 in production may harm the environment, suggesting
a productivity-environment trade-off. Furthermore,
Critical value bounds the inconsistent influence of AGRI on yields suggests
Significance I (0) Bound I (1) Bound systemic inefficiencies and the need for more consistent
10% 2.45 3.52 investment and policy. The study showed that prudent
5% 2.86 4.01 scaling of digital tools and climate-smart technology
2.5% 3.25 4.49 may decouple productivity from environmental
1% 3.74 5.06 impact, underlining the need to promote precision-
Source: Author’s estimate. based interventions, sustainable intensification, and
data-driven decision-making to boost yields and
LM serial correlation test yielded a p=0.17 and climate resilience in Pakistan through revisions of the
found no residual autocorrelation. The residuals were agricultural policy. Future research may incorporate
homoscedastic, according to the Breusch–Pagan– farmer’s behavior, financing techniques, and regional
Godfrey heteroskedasticity test with a p=0.14. The comparisons into analysis to enhance policy relevance.
Jarque–Bera normality test validated the residuals’ Pakistan’s policymakers should focus on agricultural
normal distribution with a p=0.42. The estimated sector efficiency and resource allocation to reduce
model’s stability was confirmed by the cumulative sum the detrimental effects of agricultural value addition
(CUSUM) and cumulative sum of squares (CUSUMSQ), on CCYs. To this end, robust agricultural extension
which were inside the critical bounds at 5% significance. services, sophisticated farming methods, and rural
These results suggest that the estimated ARDL model is infrastructure are needed. Policy measures that make
stable and statistically sound. high-quality inputs and new equipment more accessible,
Table 7 presents the ARDL bounds estimates, inexpensive, and simple to use should be encouraged to
revealing an F-statistic of 10.432. This value exceeds accelerate agricultural technology adoption. At this point,
the critical bounds for I(1), indicating a significant it should be noted that sustainable agricultural practices,
dependency among the model variables. Co-integration, efficient land use, adequate labor, and capital utilization
signifying a long-term equilibrium relationship between are factors that increase productivity while preventing
variables despite short-term fluctuations, was observed. environmental degradation and preserving soil fertility.
In addition, farmers must be given reliable access to real-
5. Conclusions and policy recommendation time agricultural data so that they can interpret and apply
data to make informed choices. Actions from public and
The current study showed how Pakistan’s environmental private sectors may be required to promote data analytics
conditions, agricultural production, and adoption of and precision agriculture’s digital infrastructure. Besides,
AgriTech are interconnected. The adoption of data farmers should be educated about modern farming
analytics, precision agriculture, and technology can techniques and extension services that employ sustainable
boost grain crop yields. However, the association methods and precision technology to assist farmers in
between CH and N O emissions and yields is increasing yields and reducing environmental impact.
4
2

Volume 22 Issue 3 (2025) 114 doi: 10.36922/AJWEP025130096