An Analysis Correlation between Oil Prices, Exchange Rate and Imported Inputs of Poultry Industry in Iran: Using Vine-Copula Approach

Thepoultry industry, as sub-sectors of the agricultural sector,isone of the economic activities as considered risky and play a significant role in the public life of our community. The poultry industryin Iran has a bilateral relationship with global markets because on the one hand is exporters of agriculture production and on the other hand is a major importer of inputs such a cornandsoybean. So in terms of high transactions volume, poultry industriesare influenced by international prices and volatilities. Crude oil is one of the most important commodities in the global economyand in Iran has a comparative advantage that is seen as a strategic resource. A significant portion of Iran's revenue is from oil exports account and crude oil price. Therefore, oil prices in the world is an important factor that affecting the ability of our import volume. Recent observations show that the volatility and uncertainty in oil prices are transmitted through exchange rate (USD America) to real economic markets, other markets, the exchanges and the domestic agricultural and food products markets. This articleseems clearly impressive after Iraq-USA war in 2002 and the Global Financial Crisis in 2005. So in this paper, we try to analysis correlation between oil prices, exchange rates and the price of poultry inputs for the two periods, before the Global Financial Crisis and Iraq-USA war (1995-2004) and after that (2005-2014).
Material and Theperiods ofstudy are pre and after the Iraq-USA war and the Global Financial Crisis. Our monthly data collected from the Central Bank of Iran, Animal Support Company since 1995 to 2014. For the purpose of this paper, we used Vine Copula-MGARCH approaches. Before everything at first, we controlled the stationary and seasonal unitroots behavior in data with ADF,KPSS and HEGY stationary and seasonal tests.After that for analysis the correlation of prices, we used MGARCH models for modeling volatilities and collecting the residual of equations. Because of the limitation in linear correlation coefficients and the advantages of copulas for modeling and analysis correlation, we used copula approaches for this sector. At first, we modeledvolatilities with kind of MGARCH models such as CCC and DCC GARCHes and after that for collection pure residuals we must eliminate the past effect of each variable or in other means we can tell, using ARMA with MGARCHmodel can give us residuals that have not any effect of past behaviors in variables.
The results of the ADF unit-root test has indicatedthat all variables are not stationary and accumulated from the first stages. Similarly, the KPSS unit root test has shownsuch as ADF test results. Based on these tests our variables are not stationary and in two periods of study and the first stage of a difference,they are accumulated. Seasonal unit root test or "HEGY test" results also showed there arenot seasonal behaviors in two periods of variables. After these tests for modeling volatilities at first, we needed to detect ARCH behaviors in variables. Because of that, we controlled ARCH effect behaviors in variables and for this aim, we use an ARCH-LM test. Detecting ARCH behaviors help us to use the kind of MGARCH models for modeling volatilities. Our results indicate that CCC and DCC models with ARMA model have flexibility for modeling. So after that examination, we have collected the residuals of equations and collected the residuals of each equationin ARIMA-CCC-MGARCH model. We calculated the correlation of oil price, exchange rate and input prices with kind of Vine-Copulas. Results of R-Vine, C-Vine and D-Vine models indicated that the correlation between oil price and exchange rate are different in two periods, as the positive correlation of oil and exchange rate, in the first period, change to a negative correlation in the second periods. Correlation of oil price and input pricesin second time are more than beforethe crisis. Clarke and Voung tests for choosing Vine models indicate that R-Vine models for after and before period are the best.
Based on R-Vine models our results indicated that correlation between oil price and input prices are more than before the crisis and this is not a suitable situation for Iran's industries. At last, we offer that, using oil incomes forincreased infrastructures of input productions it may be better than importing inputs.