Page 231 - Contributed Paper Session (CPS) - Volume 3

CPS2002 Atina A. et al.
Losses are said to occur if the actual yield is smaller than the estimated
yield at a certain confidence level. Otherwise, if the actual yield is greater than
the estimated yield at a certain confidence level, then farmers experience
profits.
Furthermore, value at risk in terms of losses is estimated as a reference for
farmers, on how much yield the farmers are prepared to experience losses at
certain probability level. The value at risk is calculated based on the marginal
distribution of the Y obtained from the simulation of the copula models,
ˆ
where the basic formula is
P( Y  p ) =1 − p (13)
where  is the value at risk at the 100p level.
%
p

3. Empirical Result
The procedure explained in the previous section is applied to the rice
production (hg/ha) and temperature change ( C) data of Indonesia from 1961
o
to 2016. The data obtained from the official website of BPS-Statistics
Indonesia. The rice production and temperature change data are presented in
Fig. 1.
Based on Fig. 1, rice production data has an increase trend with a little
fluctuation. While temperature changes data is quite fluctuating. To determine
whether the two variables have relationship or not, the value of Kendall’s 
and Spearman’s  are calculated. The results show that the value of each

association measure is 0.670997 and 0.852851, respectively. It means that the
two variables have quite strong dependency.
The next step is to fit the distribution of each variable. Descriptive statistics
of rice production and temperature change along with the normality test using
Shapiro-Wilk test is presented in Table 1.

Figure 1. Rice Production and Temperature Change in Indonesia from 1961
to 2016

220 | I S I W S C 2 0 1 9