Multinomial logistic regression

LR for more than two groups is often called multinomial logistic regression. The R functions mentioned in Section 5.2.3 cannot be applied in this case. Below is the outline for the multiple group case. 

In a study of performance measurement related to lean manufacturing that affects net profit of SMEs in the manufacturing sector in Thailand, the researchers used factor analysis to find the factors. The study extracted factors using the principle component method and forecasted identify factors that affect net profit by using multinomial logistic regression analysis and grouping patterns of business operations of SMEs with statistical analysis. 

The aim of this study was to evaluate the relationship between the first 2 years clinical response (qualitative analysis of proximal bone remodeling) and S independent biomechanical factors such as patient age (A), implant position angle (IPA), pre-operative bone quality (BQ), body mass (BM) and height (H). We reviewed clinical results of a consecutive series of 103 patients operated between 1989 and 1998. Classification Tree and a Multinomial Logistic regression model (Systat.ll) were used to retrospectively analyse the clinical results. 

Two multivariate analysis of the relationships between 5 biomechanical factors A, IPA, BQ, BM, H and the dependent variable CR were done using Classification Tree (CART, Systat.ll) and a Multinomial Logistic regression model (Equation. 1, Systat.ll). Two tailed p-values of 0.05 or less was used to evaluate statistical significance of predictor variables (a, b, c, d, e and constant in Equation 1). 

---