1.   Logistic regression models were used to determine which variables significantly and independently could predict bacterial pneumonia and to adjust for potential confounders.

2.   Time trends and age at diagnosis were evaluated with a linear regression model by the method of least squares.

3.   Matrix modelling extends the multiple regression model by taking into account other factors that can influence bond prices apart from term to maturity.

4.   This correlation is confirmed in the regression model which also shows a negative correlation with education level of project leader.

5.   These factors were also incorporated into a stepwise logistic regression model to identify independent risk factors.