One of the most common problems faced by analysts of agribusiness markets is that available data are aggregated to a degree that obscures the underlying decision process. This article reminds analysts of the implications of temporal data aggregation for market identification and its effects on the robustness of empirical results. Also, three major commodity market price series are analyzed to demonstrate how aggregation can affect empirical results. Finally, guidelines are suggested for selecting the appropriate level of aggregation for empirical problems.One of the most common problems faced by analysts of agribusiness markets is that available data are aggregated to a degree that obscures the underlying decision process. In particular, temporal data aggregation is a major source of specification error in economic time series analysis because it involves missing information. l,* The implications of time series aggregation for economic modeling has been occasionally analyzed since the 1950s. Yet, empirical studies of agricultural markets continue to use data sets containing daily, one day of the week, weekly average, monthly average, quarterly, and annual prices while largely ignoring the implications of such aggregation.Learner3 suggests the addition of two words to econometric discourse: "whimsy" and "fragile". According to Learner:In order to draw inferences from data as described by econometric texts, it is necessary to make whimsical assumptions. The professional audience consequently and properly withholds belief until an inference is shown to be adequately insensitive (not fragile) to the choice of assumptions.Hypothesis tests using regression often require a large number of distributional assumptions in order to estimate the models. The concept of "fragile" refers to This is Giannini Foundation Research Paper No. 906.
The authors are faculty members oft respectively,