Hi,
I am working on a report to examine weight gain outcomes in a hospital. I want to estimate the number of days it would take for children to stay in a particular type of hospital (given other factors) to gain the target weight/ WHZ score during discharge.
Please help me with suitable methods that I can use to estimate the number of days the child has to stay (preferably linear extrapolation).. Please also suggest how it is to be done in SPSS or r ?
Dear Marian, a model is supposed to interpolate and extrapolate. In the case of linear models, it is a question of extending the line. But generally, just apply the resulting model or equation to obtain the value of the dependent variable at any point of the independent variable.
It is highly questionnable that the process you describe is linear as it will asymptotically approach the "normal" weight. You first need to figure out the non-linear relationship and find an appropriate error distribution. Then run a nonlinear model on an individual child's observed trajectory. Use the parameter estimates to extrapolate.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression