Laserfiche WebLink
<br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br />I <br /> <br />.. ,.. 1'1 !'\ ~ lf~ 1:>' <br />'l.)v~.JuU <br /> <br />tial representations of real ity based on the data util ized and the <br />assumptions made. The strengths of the models are that they force the mod- <br />el builder to clearly specify assumptions which may be overlooked in more <br />sUbjective projection techniques; The model builder may detect important <br />variables that might be ignored, and the model provides a means to perform <br />"what if" analyses. The weaknesses of model s are that they may leave out <br />important factors that affect the real world, they may reflect the biases <br />of their builders, and the unrealiability of some data may make the con- <br />struction or running of the model questionable. However, after weighing <br />these strengths and weaknesses, (and because of the number and complexity <br />of the contributing factors), the use of model s has become the accepted <br />method of researchers projecting variables such as commodity prices. <br /> <br />D. BACKGROUND TO APPROACH <br /> <br />Numerous agricul tural price model s have been constructed by various <br />researchers, each model having its advantages and disadvantages. One early <br />candidate for use in the High Pl ain Study was the grain-oil seeds- livestock <br />(GOL) model developed at the USDA. Al though well documented in a three <br />volume publication, further investigation indicated that it would not be <br />the best model to use in the study. Two of its major drawbacks were that <br />it was not developed primarily as a commodity price projector and that the <br /> <br />1-4 <br />