Search

In this research we estimate the effect of El Nino Southern Oscillation (ENSO) over time on market dynamics for eight major vegetable oil prices. We estimate a system for vegetable oil prices by using a smooth transition vector error correction model (STVECM) to analyze impacts of ENSO events on production, and, more interestingly, their asymmetric nature. The results of estimated Exponential STVECM and Quadratic STAR models, respectively for the system of oil price equations and the ENSO variable regressions, suggest a smooth transition between ENSO regimes, and provide a better overall fit to the data than do linear models. Effects of the ENSO shock are analyzed using generalized impulse-response functions (GIRFs). The non-linear nature of these shocks is apparent, as the GIRFs are observed to be asymmetric depending on whether ENSO shocks are positive or negative. For most vegetable oil prices an ENSO shock has a permanent effect, meaning that prices advance to a new equilibrium level. Generally, a positive ENSO shock results in increased prices, and the opposite is true for the negative ENSO shock. The magnitude of the price change is largest for the coconut oil and palm kernel oil, and is the smallest for the ground nut oil. Also, it takes approximately two years for prices to stabilize at a new equilibrium level after the shock.

In recent years the theoretical restrictions of consumer demand have been examined in post sample forecasting exercises. However, this work has uniformly ignored the concavity restrictions of consumer demand. In this paper we evaluate a series of Normalized Quadratic Inverse Demand System (NQIDS) specifications using rolling windows and generating one-- to four--step ahead forecasts. To estimate the models, eleven categories of South Atlantic fish are used from 1980 through 2001. In addition to the NQIDS, we also examine the forecasting performance of a purely time series model. We find that the best predictions are achieved using a composite forecast.

In recent years the theoretical restrictions of consumer demand have been examined in post sample forecasting exercises. However, this work has uniformly ignored the concavity restrictions of consumer demand. In this paper we evaluate a series of Normalized Quadratic Inverse Demand System (NQIDS) specifications using rolling windows and generating one-- to four--step ahead forecasts. To estimate the models, eleven categories of South Atlantic fish are used from 1980 through 2001. In addition to the NQIDS, we also examine the forecasting performance of a purely time series model. We find that the best predictions are achieved using a composite forecast.

We extend Holt and Bishop's (2002) normalized quadratic inverse demand function by obtaining measures of precision of elasticities and consumer welfare loss estimates due to reductions in commercial landings in the U.S. Great Lakes. Even with curvature constraints imposed, confidence intervals are obtained by using the bootstrap, subsample bootstrap, and subsample jackknife., Replaced with revised version of paper 06/30/04.

In this paper we introduce inverse demand systems that include quadratic scale terms. These systems are similar to regular quadratic demand systems introduced by Howe, Pollack, and Wales. A unique feature of these specifications is that they maintain linear scale curves as a special case. For illustrative purposes we estimate the Normalized Quadratic Inverse Demand-Quadratic Scale System using monthly South Atlantic fish landings and valuation data, 1980-1996. In estimation concavity is maintained locally, and the rank reduction procedures advocated by Diewert and Wales (1988b) are employed. The estimated model is then used to obtain welfare estimates associated with catch restrictions.

The impacts of introducing a partial price stabilization scheme in the U.S. corn market are investigated by using a modified version of the bounded price variation model. Specifically, a model is developed and estimated that includes rational expectations of the first three central moments of the (truncated) equilibrium price distribution. The estimated model is used to stimulate market equilibrium effects of introducing upper and lower price limits through a tax-subsidy scheme. The results show that corn producers are downside risk averse, and that market feedback effects of price stabilization can, at times, be more important than direct effects.

The role of price risk in sow farrowings is investigated by using bivariate ARCH-M and GARCH-M models and a nonparametric kernel estimator. To account for the relevant time horizon of irreversible supply decisions, predictions for mean price and conditional price variance are iterated forward. The empirical results vary markedly in terms of their implications for risk response in hog supply decisions, with the ARCH-M and GARCH-M models suggesting a small and negative risk effect. Estimates of the marginal risk premium also indicate moderate and variable departures from marginal cost pricing in sow farrowing supply decisions.

This paper extends the bounded prices model under rational expectations to a multi-market setting. The resulting framework is used to estimate a supply-demand model for corn and soybeans. The estimated model is used to simulate the implications of removing price support and diversion programs over the sample period.

This article applies recent developments in time-series modeling to analyze the retail prices of beef, pork, and chicken. Specifically, generalized autoregressive conditional heteroscedasticity (GARCH) models were fitted to these data to determine if, unlike more traditional time-series models, the conditional variances of the underlying stochastic processes are nonconstant. The estimation results indicate that the constant conditional variances assumption can be rejected. Furthermore, ex post forecast intervals generated from the GARCH processes indicate that the forecasting accuracy of the estimated models has varied widely over time with substantial volatility occurring during the 1970s and early 1980s.