3. Results and Discussion
3.1. Results from VAR Model: The results of the VAR model
estimation for the key variables of interest, including wheat production
(Wheat), carbon dioxide concentrations (CO2), average temperature
(temperature), average precipitation (Precip), and agricultural land
used for wheat cultivation (Area), and water accessibility (Water) from
SPSS, are shown in the Fig.1 below. The
analysis of the relationships between environmental factors and
agricultural production involved the application of the Vector
Autoregression (VAR) model. This model encompassed five variables,
namely Area used for wheat cultivation, CO2 emission, Precipitation,
Temperature, Water, and Wheat yield. To estimate the VAR model, a sample
of 50 observations from the year 2022 was utilized after making
necessary adjustments. The model demonstrated a strong overall fit, as
indicated by the R-squared values ranging from 0.90 to 0.99 for
different variables. The F-statistics confirmed the joint significance
of the variables in each equation, with values ranging from 2.94 to
16.76 (Table 1). The Akaike Information Criterion (AIC) and Schwarz SC
indicated a relatively good fit for the model, with lower values
suggesting better fit. Specifically, the AIC value of 16.70483 and SC
value of 16.97509 for the lag 1 model indicate a higher level of
parsimony. As a result, the VAR model with a lag of 1 is thought to be
better for this inquiry than other lag values (See Fig 1).
Furthermore, the Adj. R-squared values reinforced the model’s
explanatory power. The goodness of fit was reflected by the sum of
squared residuals provided for each variable. The equations’ standard
errors varied from 0.98 to 960.72, highlighting the precision of the
model. Overall, the VAR model revealed strong relationships among the
variables under study, demonstrating high explanatory power and a
satisfactory fit. The coefficients offered valuable insights into the
dynamic interactions within the environmental and agricultural system.
Using this model, the wheat yield for the year 2030 was estimated.
3.2. Prediction of Wheat for 2030:The calculation below is used to predict the anticipated value for wheat
yield in 2030 using the VAR technique with a lag of one:
E (Wheat 2030) = –7210.404 + 0.186449 (wheat2022) + 0.131691 (CO22022) + 265.6333
(Avg. Temp 2022) + 16.29369 (Avg. Precip 2022) +
95.77185 (Water 2022) + 0.028147 (Area2022)
E (Wheat 2030) = –7210.404 + 0.186449 (24033)
+ 0.131691 (48174) + 265.6333 (22.6) + 16.29369 (39.2) + 95.77185
(142.9) + 0.028147 (9046)
E (Wheat 2030) = 24197.09
Based on our calculations, the projected wheat output for the year
2030 is expected to be 24197.09 thousand tons. However, the official
government number obtained from the (Economic Survey, 2022) reports the
actual wheat production in 2030 as 23864 thousand tons.