Search

We extend the distributional event response model (DERM) of Rucker, Thurman, and Yoder (2005) in two ways. First, we develop a mixed event response model (MERM) to allow for possible asymmetric effects, and second, we examine how volatility, in addition to return, changes surrounding an event. We apply our model to the crude oil futures market using 25 years of daily data. Our results show that among the 10 events considered, the 2008 global financial crisis had the largest impact in magnitude on both return and volatility. The location and duration of response patterns are also found to vary across different events, with the financial crises having long-lasting impacts, while truly unanticipated events, such as the September 11 terrorist attacks, having short-lived impacts. Results suggest that simply using an event-day dummy variable would hinder discovering overall market responses to slowly evolving information events.

Improving rural broadband access has been touted as a rural development strategy, but there is limited evidence that broadband service affects rural economic growth. We measure the effect of broadband deployment on location decisions of new rural firms. Location-specific fixed effects are controlled by a counterfactual baseline that measures how local broadband service in the early 2000s affected local new firm entry in the early 1990s before broadband was available anywhere. The change in location choice probability of new firms from the counterfactual baseline to the actual response ten years later is the difference-in-differences estimate of the effect of broadband deployment on locations of new firms. We find that broadband availability has a positive and significant effect on location decisions of new firms in rural areas, which is confirmed by a robustness test using ZIP Code dummy variables. The broadband effect is largest in more populated rural areas and those adjacent to a metropolitan area, suggesting that this effect increases with agglomeration economies.

In this article we estimate a state-contingent production frontier for a group of farms while endogenously estimating the number of states of nature induced by unobserved environmental variables. This estimation is conducted by using a birth-death Markov chain Monte Carlo method. State-contingent output is estimated conditioned on an observed input vector and an a priori unknown number of unobserved states, each of which is modeled as a component of a mixture of Gaussian distributions. In a panel data application, state-independent dummy variables are used to control for time effects. The model is applied to 44 rice farms in the Philippines operating between 1990 and 1997. The endogenous estimation procedure indicates a unimodal posterior probability distribution on the number of states, with a median of three states. The estimated posterior coefficient values and their economic implications are compared to those of previous research that had assumed a fixed number of states determined exogenously. Goodness-of-fit testing is performed for the first time for a state-contingent production model. The results indicate satisfactory fit and also provide insights regarding the degree of estimation error reduction achieved by utilizing a distribution for the number of states instead of a point estimate. All of our models show significant improvement in terms of mean squared error of in-sample predictions against previous work. This application also demonstrates that using a state-independent dummy time trend instead of a state-contingent linear time trend leads to slightly smaller differences in state mean output levels, although input elasticities remain state-contingent.

New residents of rural communities are often assumed to have preferences for development and conservation that differ from those of longer-term residents. However, the literature offers little to quantify presumed preference heterogeneity. This article assesses whether stated preferences differ according to length of residency. Results are based on a conjoint (choice experiment) survey of Rhode Island rural residents. Heterogeneity—according to length of town residency—is modeled using dummy variables, multiplicative interactions, and Lagrangian interpolation polynomials. Results are compared across the three models, and identify a range of attributes for which willingness to pay depends on length of residency. Copyright 2003, Oxford University Press.

This study compares the extent of technology use by rural and urban manufacturing plants. Count data regression models explain technology use as a function of rural-urban indicator variables, plant characteristics, and location characteristics. The results show no difference in technology use due to rural-urban location. Characteristics of nonmetro plants, including size, prevalency of branch plants, and nature of production process, tend to raise nonmetro manufacturing technology use. Lower nonmetro schooling levels, an older labor force, and mix of industries with low technology use tend to reduce technology use in nonmetro areas. Copyright 1998, Oxford University Press.

Nerlovian supply functions are fitted to post-World War II coffee data for São Paulo, Brazil. Supply shifts are permitted through the use of dummy variables, and recent specifications of irreversible functions are tested. The modified Nerlovian functions perform best from both the economic and statistical point of view.

Animal production experiments permit limited control of experimental factors because the animals themselves partially determine total feed consumption per unit of time. Within practical limits, the variables that economists would like to have as controlled factors in the experiment are not subject to direct control, but instead, are controlled indirectly by other closely related variables. The problems involved are illustrated by a recent article of Sonka, Heady, and Dahm. In a swinefeeding experiment, percentage of protein in the ration was the only design variable, but economists would have liked to see corn and protein supplement as the controlled factors since they are the direct inputs for which there exist price quotations. Of course, other variables besides percentage of protein can be thought of as the single controlled factor; either percentage of the ration by weight which is corn or a standardized protein supplement would serve equally well. The important point to be recognized is that the controlled factors are different than the direct feed inputs and one less in number because the animals themselves control total feed consumption per unit time. In the swinefeeding experiment analyzed in Sonka, Heady, and Dahm, percentage of protein is the single factor instead of the pair of variables, corn and protein supplement consumption. Data from the swine-feeding experiment can be analyzed in two basic ways: (a) weight gain, corn consumption, and protein supplement consumption per unit of time are treated as multivariate responses to the single controlled factor, namely, percentage of protein in the ration; and (b) weight gain is divided into discrete intervals, and for each of these intervals, corn consumption, protein supplement consumption, and time required to achieve a specific gain are treated as multivariate responses to the single experimental factor. Following (Sonka, Heady, Dahm), we consider only the latter type of analysis which simplifies statistical estimation problems. Adopting most of the notation in Sonka, Heady, and Dahm, there are three equations of possible economic interest to be estimated from the experimental data

Apparently Teague has not reviewed all of the literature for examining relative efficiency. In the early work of Lau and Yotopoulos (LY), differences in economic efficiency are tested by incorporating only a dummy variable in the profit function to quantify differences associated with particular groups. (Teague implicitly argues that only this should have been done as evidenced by the assertion that equations (2) and (3) are redundant.) In an extension of this early work, Yotopoulos and Lau (YL) expand their conceptual framework to permit the assessment of relative technical and pricing efficiency as well as overall economic efficiency. This latter framework is utilized by Garcia, Sonka, and Yoo. Under this formulation, the profit functions of the two firms or groups under analysis differ by a constant factor which is a function of technical and pricing efficiency. In addition, the demand equations may differ across groups by constant factors which are a function of pricing efficiency. Similar formulations also are presented in Trosper, O'Connor and Hammonds, and Khan and Maki. Perhaps Teague's difficulty arises from either a misunderstanding of the methodology or an unfamiliarity with the approach and its standard notation. In the former case, the methodology permits tests for the relative efficiency (pricing and economic) without imposing it. That is, actual UOP and demand functions can be derived based on behavioral relationships. Differences in actual functions permit an assessment of relative efficiency. In the latter, producers respond to "effective prices." Profits and demand equations are derived based on these effective prices, not actual prices. The p's in Garcia, Sonka, and Yoo can be shown to represent an underlying structure that permits differences in efficiency between the groups based on effective prices. To understand the logic behind this formulation, consider the concept of relative efficiency developed by YL. This approach begins with the notion that given comparable endowments, identical technology, and normalized input prices, the actual profits of two firms should be identical if they both maximize profits. To the extent that one is more price efficient, or technically more efficient, than the other, the normalized profits will differ even for the same normalized input prices and endowments of fixed inputs. The actual normalized profit functions will then be different. Consider the case of the production functions of two firms: (4) Y = AIF(X', ZI); and r2 = A2F(X2, Z2), where the X's and Z's are variable and fixed inputs, respectively, A1 and A2 are scalar technical efficiency parameters, and the superscripts represent the two firms. The marginal conditions can be represented by

Policy variables are typically treated as exogenous variables in agricultural commodity models. In this paper evidence that policy variables can be successfully endogenized in a world trade model of the soybean/rapeseed and products market is presented. Minimum price support equations are estimated for U.S., Brazilian, and Japanese soybeans, and EC and Japanese rapeseed. Macroeconomic indicator variables are used to capture the impact of nontariff barriers in Brazil and Japan domestic-world price link equations. Simulation experiments are conducted which indicate that the model with endogenous policy variables performs well in comparison with a model which treats policy exogenously.

For years agricultural economists have been concerned with obtaining reliable estimates of production function parameters. Among other reasons, such estimates are needed to address policy issues involving resource misallocation, supply response, and factor demand in agriculture. The appeal of production function approaches to such problems, as contrasted with direct estimation of product supply and factor demand relationships, is with the richness of the model for addressing cross-price effects and matters of resource allocation. However, as is well known, problems of multicollinearity and technological change have thwarted most efforts to obtain reliable estimates of aggregate production elasticities by direct estimation (Doll; Rosine and Helmberger). In an effort to circumvent the multicollinearity barrier posed by direct estimation, Tyner and Tweeten (1965) indirectly estimated aggregate production elasticities for U.S. agriculture using a factor share approach. Noting that in equilibrium, factor shares equal factor elasticities for an industry comprised of profitmaximizing firms, Tyner and Tweeten synthesized a Cobb-Douglas production function by estimating equilibrium factor shares based on historical data and the Nerlove partial adjustment model. Equilibrium factor shares were estimated individually for each of nine input categories assuming that actual adjustment between any two years was a constant proportion of the extent of own-factor disequilibrium. Estimates were developed for several decades by including dummy variables in the model. In a subsequent article Tyner and Tweeten (1966) developed estimates of resource misallocation, product supply elasticity, and factor demand elasticities based on their indirectly estimated CobbDouglas production functions. In 1974 Rosine and Helmberger published an article with a similar intent. They also explored several macro policy issues based on a synthesized Cobb-Douglas production function for U.S. agriculture. They used a simpler model than Tyner and Tweeten, by assuming that production elasticities were given by actual factor shares. In effect, Rosine and Helmberger assumed instantaneous adjustment to equilibrium thereby avoiding numerous estimation difficulties. (See writings of D. Gale Johnson in 1948 and Ruttan and Stout in 1960 for examples of earlier expressions of interest in factor shares.) To develop normative estimates of production elasticities following a factor share approach, it seems clear, a priori, that the Tyner-Tweeten approach should be preferable conceptually to that of Rosine and Helmberger. In the words of Tyner and Tweeten, "the use of [actual] factor shares as production elasticities trades the intercorrelation problem for the dubious assumption that economic equilibrium prevails" (1965, p. 1462). The purpose of this paper is to evaluate the practical effects of the assumption of continual economic equilibrium (instantaneous adjustment) on production elasticity estimation.

The purpose of this paper is to examine the relationships between feed grain prices and the beef product mix. An annual econometric model of the U.S. livestock industry (Ospina and Shumway 1979, 1980) is used to evaluate the impact of corn price changes on the composition of slaughter beef supplies and the relative prices of the quality categories by analyzing the reduced-form dynamic multipliers. The structural form of this model disaggregates beef demand among three quality components (U.S. Department of Agriculture [USDA] grades prime and choice, good, and standard-and-lower, labeled for convenience as choice, good, and utility, respectively).' Supplies are disaggregated among the same quality categories and also among sex categories (steers, heifers, and breeding herd culls). Simultaneity is modeled in the demand, supply, and inventory decisions. Pork and broiler functions are incorporated. The separate effects of current and expected prices on supplies are differeptiated, and feed grain prices are assumed to be exogenous.2 All functions are linear, and the analysis is at the carcass level. Three changes were made in the earlier model, and its parameters are reestimated (Ospina and Shumway 1981). The changes include the following. (a) Three subsequent years of data are added, thus making a twenty-three-year data period, 1956-78. (b) Because important changes in USDA beef grades took place in early 1976 which affected the choice and good grade definitions (Nelson), an intercept dummy variable is added to these supply and demand equations to account for its impact. (c) A relatively minor change is made in the disaggregation of data among the good and utility beef categories.3 The change in USDA grade definitions significantly increases the supply of choice beef and decreases the supply of good beef. The estimated parameters of the grade change dummy variables have expected signs and low standard errors in five of the six structural equations to which they are added. Several substantive differences from the earlier estimation occur in the magnitudes of individual parameters because of the additional data and other modifications. However, parameter signs and the earlier general conclusions are stable. In particular, the aggregate short-run beef supply response remains positive (this finding contrasts with the conclusions of some earlier writers, e.g., Reutlinger, Tryfos), and the supply response to a 1% decrease in corn price remains greater than the response to a 1% increase in beef prices. At 1978 prices, the elasticities of beef supply are estimated to be -0.23 with respect to corn price and +0.15 with respect to average beef price. These estimates are similar to our earlier estimates of -0.25 and +0.14, respectively, at mean 1956-75 prices.

A conceptual clarification of the sources and meaning of cross-sectional price variability is used to motivate a theoretical and econometric framework for the estimation of cross-sectional demand functions. Quality effects are distinguished from supply-related price variability to identify cross-sectional demand for disaggregated food commodities. An empirical application using data from the 1977-78 Nationwide Food Consumption Survey indicates that parameter differences resulting from a failure to adjust cross-sectional prices for quality effects are likely to be small for relatively homogenous, disaggregated food commodities. In demand analysis with cross-sectional, household budget data, it is usually assumed that prices are constant (Allen and Bowley, Prais and Houthakker, George and King). Given this assumption, Engel functions are estimated where expenditure (or quantity) is regressed on income (or total expenditures), family size, and other demographic characteristics. The assumption that cross-sectional price effects are absent or are captured adequately by spatial and temporal dummy variables has not been evaluated empirically. Whether cross-sectional price effects can be treated in this manner has implications for the specification of cross-sectional demand functions as well as the estimation of Engel functions. There is a considerable literature on estimating price elasticities with cross-sectional data. Most of the applications utilize timeseries/cross-sectional data. Annual (or quarterly) household consumption surveys and

Federal milk marketing orders contain formulas that determine prices received by farmers for milk delivered to bottlers. Different orders contain various combinations of provisions. Cross-section data were used to test the hypothesis that order prices are related to the provisions contained in orders. The hypothesis was accepted. Some 78 percent of the variance in prices is explained by two economic variables: distance of order market from Eau Claire, Wisconsin, and class I utilization ratio. Inclusion of dummy variables to allow for effects of order provisions increased the proportion of explained variance to 94 percent. Covariance analysis found no significant changes between years in the effects of order provisions on prices. Many possible combinations of provisions are not used in any order. Many of these combinations would provide unacceptable prices if they were used. IN RECENT years the problem of maintaining proper milk price relationships among federal order fluid milk markets has become increasingly serious. This is mainly due to three factors: (a) increase in the size and number of orders, (b) improvements in the highway system and developments in longdistance hauling of bulk and of packaged milk, and (c) growth in size of grocery chain stores and bottling plants. The Federal Milk Order Study Committee was concerned with problems of intermarket relationships [2, pp. II-2-1 to 11-2-15, II-4-1 to II-4-56, III-19 to III-923. In 1955 the Agricultural Marketing Service published a study which dealt, in part, with intermarket price relationships [81. Later, Babb [1] analyzed intermarket price relationships. The objective of my study was to test hypotheses concerning effects of federal order provisions on relationships among prices in federal order markets. Dummy variables were used to study effects of order provisions. Covariance analysis was used to test for variation in the effects of these provisions. The hypotheses tested will be presented after a brief discussion of federal order provisions; more complete descriptions can be found in publications

Often the estimation of agricultural production technologies in less developed countries is achieved by dividing the sample arbitrarily into "small" and "large" farms and estimating a separate production function for each group. This is equivalent to including an appropriate dummy variable in a single regression on the whole sample. A modification of this approach is to replace the dummy variable (a step-function) by either a cumulative normal or logistic function. The parameters of such a "generalized dummy variable" convey valuable information on whether a small-large categorization is appropriate and, if so, where the division should be made.