Date of Award:


Document Type:


Degree Name:

Master of Science (MS)


Applied Economics

Committee Chair(s)

Herbert H. Fullerton


Herbert H. Fullerton


Roice H. Anderson


Allen D. LeBaron


Darwin B. Nielsen


The primary objective of this study was to develop demand equations at retail for selected Venezuelan food commodities. Commodities included for analysis were rice, refined sugar, crude sugar, flour, potatoes, beef, pork, black beans, corn, and powdered milk.

The basic statistical technique employed in the analysis was least squares multiple regression. Although several mathematical forms for these demand equations were evaluated, a log-log transform was found to be most useful. Independent variables included own price, prices of substitutes and complements, time and income. Observations were taken from Venezuelan time series data over the period 1945-1965.

Development of these basic demand equations facilitated the treatment of a second objective which was to evaluate the degree of complementarity and substitution between each one of the commodities included in the model. Further, it provided the necessary structural framework to give exante examination to selected policy alternatives for Venezuelan agriculture. Alternatives examined included minimwn price and import policies.

Interesting results and conclusions may be s ummarized as follows:

1. Difficult statistical problems are encountered in an attempt to estimate direct and cross elasticities of demand from time series data.

2. A less general approach utilizing principles of demand theory is advisable until further requirements are made in the data series.

3. A consistent demand model docs provide useful insights into the interrelationships of commodity demands in terms of direction of change in price and quantity if not in terms of their magnitude.

4. Regional and social stratification of price, income and consumption data should improve the reliability of any subsequent analyses.

5. Additional household expenditure surveys of a cross-sectional nature would provide useful and necessary checks on the demand estimates obtained from time series data.