Abstract:
Vector Autoregression (VAR) is an econometric approach widely applied to examine interactions
among economic variables, generate forecasts, and guide policy-related decisions. This
research primarily captured short- to medium-term dynamics between climate factors and vegetable
prices in Badulla, focusing on cabbage and beetroot within the Haputale and Bandarawela
regions from 2000 to 2024. Monthly wholesale price information was obtained from the Hector
Kobbekaduwa Agrarian Research and Training Institute (HARTI), while detailed climate data—
mean monthly temperature (◦C) and total monthly rainfall (mm)—were collected from the Department
ofMeteorology.The analysis followed three main steps. First, stationarity was verified
through Augmented Dickey–Fuller (ADF) and Phillips–Perron (PP) unit root tests, confirming
that all variables became stationary after first differencing (p < 0.01). Second, the appropriate
lag order for the VAR model was identified as four months, based on several information criteria
(AIC = 3.21, BIC = 3.45, HQ = 3.32). Third, Granger causality tests revealed significant
relationships: rainfall Granger-caused cabbage price changes (F-stat = 5.34, p = 0.003) and
temperature Granger-caused beetroot price variations (F-stat = 4.87, p = 0.008). The VAR coefficients
showed that a 1% increase in rainfall was associated with a 0.42% (SE = 0.11) price
increase for cabbage in the subsequent month, while a 1 ◦C temperature rise led to a 0.38% (SE
= 0.09) price increase for beetroot after two months. Impulse response functions demonstrated
that climate shocks had persistent effects: a one standard deviation rainfall shock (≈ 50 mm)
raised cabbage prices by 2.1 to 3.4% over six months, while an equivalent temperature shock
(≈ 1.2 ◦C) increased beetroot prices by 1.8–2.5% with a three-month lag. Variance decomposition
analysis indicated that climate factors explained 28% of cabbage price variation and 22% of
beetroot price variation at the 12-month horizon. These findings suggest that rainfall-responsive
pricing mechanisms could help stabilise cabbage markets during monsoon seasons, and early
warning systems using the VAR framework could predict price spikes with a 2-3 month lead
time. The study contributes to climate-resilient agriculture by quantifying asymmetric price
responses to different climate variables across distinct vegetable crops.
