SOME STATE SPACE MODELS OF AIDS EPIDEMIOLOGY IN HOMOSEXUAL POPULATIONS - Quantitative Medical Data Analysis Using Mathematical Tools and Statistical Techniques - page 56

(a) The estimates of (d S ; d I (u) = d I ) are given by (8:3510 7 4:49
10 7 ; 310 4 1:6210 4 ) per month respectively. These results indicate
that the death and retirement rate of I people is much greater (at least 100
times greater) than that of S people in the Swiss population of homosexual
and bisexual men. This suggests that HIV infection may have increased the
death rates of HIV infected people.
(b) The estimate of the proportion of immigration and recruitment rate
are 2:4810 3 6:09410 4 per month for the population of homosexual
and bisexual men. This estimates is about 10 times greater than that of the
estimates of the death and retirement rates of the I people. These results
indicate that the size of the Swiss population of homosexual and bisexual
men is increasing with time.
(c) From Figure 1, the estimated density of the HIV infection distri-
butions showed a mixture of distributions with two obvious peaks in the
Swiss population of homosexual and bisexual men. The rst peak occurs
around May of 1980. The second peak occurs around August of 1992 and
is considerably lower than that of the rst peak. Comparing the estimated
density of the HIV infection in Figure 1 with the estimated density from
the San Francisco homosexual population by Tan and Xiang 14 , one may
notice that the two curves are quite similar to each other but the Swiss
population appears about 6 months earlier for the rst peak and about 2
years earlier for the second peak.