FIGURE A.4 Median prices of powder cocaine and cocaine base in three cities.

where is the predicted median price and is the estimated value of βj. It can be seen from the figure that the median prices of powder cocaine and cocaine base are very different.3 There is no systematic relation between the prices of the two forms of cocaine. The price of powder cocaine was higher than that of cocaine base in San Diego during 1987–1989, but the price of cocaine base was higher in Boston during 1994–1998. In Wash

3  

An asymptotic chi-square test rejects the hypothesis that the coefficients of the models for powder cocaine and cocaine base are equal (p<0.01 for each city). In other words, the test rejects the hypothesis that the price functions of powder cocaine and cocaine base are equal. The test procedure is as follows. The hypothesis has the form H0:β(p)=β(b), where β(p) and β(b), where β(p) and β(b) are vectors of the coefficients of the models for powder cocaine and cocaine base. Let and , respectively, be the estimators of β(p) and β(b), and let and be the consistent estimators of the covariance matrices of their asymptotic normal distributions. Let d be the dimension of and . Then if H0 is true, is asymptotically chi-square distributed with d degrees of freedom. The asymptotic chi-square test rejects H0 at the α level if χ2 exceeds the 1 —α quantile of the chi-square distribution with d degrees of freedom. It is important to note that the p value of a hypothesis test based on the STRIDE data does not have its usual meaning because the data are not a random sample. The p value may, however, be used informally to indicate whether the observed value of a test statistic is likely or unlikely under repeated sampling of a population whose distribution is similar to the empirical distribution of the data.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement