DIGITALNA ARHIVA ŠUMARSKOG LISTA
prilagođeno pretraživanje po punom tekstu
| ŠUMARSKI LIST 3-4/2018 str. 28 <-- 28 --> PDF |
The results of the regression analysis are shown in Table 3. The goal was to identify the best linear model for prediction of the GDD values for each of the GDD groups. First, univariate linear regressions were calculated for all of the analysed variables (GDD was the dependent variable; Table 2). From the results for all three groups, insolation and day of year are clearly statistically significant predictors for GDD. Although these factors are significant predictors, their coefficients of determination range from 71 % to 89 % for insolation and from 26 % to 62 % for the day of year. In addition to the univariate models, we created a multivariate model for the variables were shown to be statistically significant in the univariate models, i.e., insolation and day of year. When we include the value of the day of year to the insolation value for the early flushing group of Pedunculate Oak, |