Multivariate linear regression is used to examine the possible effects of the urban air pollutants sulphur dioxide
(SO2) and particulate matter (PM10) on the number of asthma cases at the district level in Izmir between the years
2007 and 2010, taking into account the effect of the total population. The selected modelis a no-intercept model. The
selected model explains 88% (R2=0.88) proportion of the variability in the recorded asthma cases. Remark that the
selected model has no constant, so the R2 cannot be compared with the models with a constant. Dummy variables
are used to detect the effect of the district and the year but these variables did not yield statistically significant
results. The parameter estimates are presented in Table 5.