International Journal of Applied Research
Vol. 2, Issue 9, Part D (2016)
Cost analysis of alternative modes of delivery by lognormal regression model
By investigating relationship between response variable cost of delivery & explanatory variables mode of delivery, type of institution, area of residence, this study compares the estimates of response variable cost of delivery by linear multivariate regression analysis & log-linear multiple regression model. When the linear multivariate analysis is carried out the estimated value for explanatory variable is negative & significant which seems to be unrealistic. Even though by removing outliers, the results do not change. When log-linear multivariate analysis is carried out the estimates are positive & significant hence reliable. By comparing coefficient of determination R2, it is concluded that log transformed data is more suitable than the conventional multiple regression as coefficient of determination is higher for log transformed data.
How to cite this article:
Sunanda Tanaji Wagh. Cost analysis of alternative modes of delivery by lognormal regression model. Int J Appl Res 2016;2(9):215-219.