Here is an example from ext.nodak.edu
Empirical “bioeconomic” models, which include information on biological
relationships, economic relationships, and interactions between them, are a useful tool for
policymakers addressing an invasive species problem (Eiswerth and Johnson, 2002;
Knowler and Barbier, 2000). The usual concerns apply when evaluating invasive species
policy using estimates or predictions based on data collected in the absence of invaders—
care must be taken in interpreting data summarizing producer and market behavior,
because policy decisions affect management decisions.
To illustrate this general point, this paper reports on a case study of a whitefly
invasion of California strawberry fields. We evaluate two alternative methods for
modeling whitefly population dynamics: a reduced-form autoregressive econometric
model and a structural, calibrated simulation model. The reduced-form autoregressive
model uses statistical techniques and historical data on the pest population to predict the
future population. It is a simple approach and requires limited information on the pest
population. This may be an attractive option for policy makers in the case of a biological
invasion, since limited data will still permit rapid policy analysis. Such a model,
7/6/2006
however, may omit important biological factors, such as variations in the population
growth rate over time, if no statistical data are available. The model’s effectiveness is
also dependent on the validity of the imposed statistical assumptions.
On the other hand, a simulation model can incorporate data obtained outside a
statistically valid scientific experiment. For example, it can integrate results from studies
of the pest in similar environments with results from experiments regarding the current
invasion. However, the conclusions that result depend on how biological and economic
relationships are specified, whether obtained from outside sources or assumed.
Our analysis compares the estimated cost of a pesticide use regulation obtained
using these two modeling approaches. In this instance, incorporating all available
information and constructing a structural simulation model, rather than limiting attention
to estimating a reduced form autoregressive model, results in a lower estimate of the cost
of the regulation. Because this estimate reflects information known about the whitefly
from other sources, but not available in our experimental data, it is more likely to be
correct than the reduced form autoregressive model is. The difference in our two
estimates is not itself a measure of the value of biological information, but the fact that
the two estimates differ indicates that such information could be quite valuable. This is
particularly true in circumstances where use of an incorrect measure of a policy’s cost
might have created an incentive to modify the policy in order to reduce its cost.
2007-03-19 15:28:50
·
answer #1
·
answered by Santa Barbara 7
·
0⤊
0⤋