Log in
RegisterTo diff or not to diff, that’s the question [PK / PD]
posted by Helmut 
– Vienna, Austria, 2012-01-13 16:20 (5272 d 13:13 ago) – Posting: # 7926
Views: 7,237
Dear Martin!
Shame on me. Pointed out by John Wagner already in the 1970s… Actually I started with the residual plots and sneeked into the AICs later. Will never do it again. Promise.
Agree about the basal levels. I just wanted to advise against nasty questions arising concerning volumes of distribution and clearances, like “Whaaat did you say are a1 and a2?” If you are not comfortable with Laplace transforms you can go with a symbolic maths package (my favorite: Maxima; moderate learning curve). If you obtain an explicit solution from the system of differential equations you can fit the hybrid constants and backtransform them later on.
❝ Regarding the AIC: according to Gabrielsson and Weiner (page 460) the AIC or SC (Schwarz criteria) cannot be used to compare the models as different weighting schemes are used.
Shame on me. Pointed out by John Wagner already in the 1970s… Actually I started with the residual plots and sneeked into the AICs later. Will never do it again. Promise.
❝ […] I think an estimate of the basal effect obtained by the above model is better understandable to non-experts than the turnover rate estimated via Johan's model (personal opinion/experience).
Agree about the basal levels. I just wanted to advise against nasty questions arising concerning volumes of distribution and clearances, like “Whaaat did you say are a1 and a2?” If you are not comfortable with Laplace transforms you can go with a symbolic maths package (my favorite: Maxima; moderate learning curve). If you obtain an explicit solution from the system of differential equations you can fit the hybrid constants and backtransform them later on.
—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png)
Helmut Schütz
![[image]](https://static.bebac.at/img/CC by.png)
The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz
The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Complete thread:
- endogenous compounds martin 2012-01-11 21:02 [PK / PD]
![[Mix]](https://static.bebac.at/img/mix.png "open in Mix view")
- To diff or not to diff, that’s the question Helmut 2012-01-13 02:06
- To diff or not to diff, that’s the question martin 2012-01-13 09:19
- To diff or not to diff, that’s the questionHelmut 2012-01-13 15:20
- nasty questions martin 2012-01-13 15:55
- quick answers Helmut 2012-01-13 16:09
- nasty questions martin 2012-01-13 15:55
- To diff or not to diff, that’s the questionHelmut 2012-01-13 15:20
- To diff or not to diff, that’s the question martin 2012-01-13 09:19
- To diff or not to diff, that’s the question Helmut 2012-01-13 02:06
Ing. Helmut Schütz