Talk:Disability-adjusted life year
Editing/adding to wiki articles based on other documents
I just added a couple of lines to this article and upon adding a reference section at the bottom, noticed that the page was in fact a copy of a document available elsewhere as a report. It made me wonder - should we not edit pages like this in the INTARESE-wiki? Or should we edit/add but remove such 'source' tags in the articles? Any thoughts? BTW If anyone disagrees with changing such articles, please feel free to revert to an earlier version. Jgrellier 18:28, 24 April 2007 (EEST)
Discussions on developing DALY function/tool
Template for checking out Analytica files is now available. Instruction can be found from Help pages -> Locking Analytica files
Dear Anne, Marko and Olli,
I'm delighted to see cooperation on DALYs, and especially when it is done in Analytica. Further, I'd like to suggest that we/you could work on this in wiki in the following way:
- I uploaded the model (in XML format) to Interese wiki: Image:Disability-adjusted life year.ANA. This way, the model is available to everyone interested in playing with it.
- When you want to EDIT the model, do the following:
- Download the newest version.
- Make an edit to the file description page and add the template Template:checkout.
- When your edits are complete upload the edited version WITH THE SAME NAME (use "Upload a new version of this file" button).
- Remove the template from the file description page.
- The description of the model functionality should be in the Disability-adjusted life year page.
- The discussion and questions about the model functionality should be in the Variable Talk:Disability-adjusted life year page (this page).
Then some comments on the model itself (in addition to what Marko said already):
- The beauty of Analytica is that you can chop things into modules, and then think about things within modules. This makes life easy when you get used to it. For example, index "Year" should not be used here at all. If year is an important thing in the upstream values such as exposure, the calculations and results will automatically generate the right kind of table indexed by the year.
- Tables as definitions of formulae should be avoided. Tables are good for upstream data, but not for calculations (although there are exceptions such as the FRES model where tables are used to describe how a large number of details are agglomerated into a huge result table). Conclusion: if you think you need a table for formulas, write what you want to do on this page, and we'll tell you how to do it.
- With severe annoyance and sleep disturbance: It seems that the same formula (or function) is repeated several times, but with different numbers (input parameters). If this is the case, I would suggest: Make one variable that contains the formula, and another one with the parameter values. This way, it is easy to discuss the parameters separately, and you can easily change the values without a need to go into the formulas.
- As a general rule: try to separate the functions and the parameter values into different places. Usually it clarifies things and directs your attention into right questions.
- I did some restructuring of the noise module, but I think the two Indexes used (Lden, Lnight) are inconsistent. What is their purpose? In addition, the third index Letmaal is not used at all. Should it be removed?
--Jouni 11:15, 14 June 2007 (EEST)
I should not have wrote on good weather for you. It brought bad luck (and weather)... Well, some rain is good for plants.
I took a look to your model and I also took liberty to change some of the codes on it... The new version is attached. Now I think the model calculates what it should calculate. Few remarks: - We use 0 values if there is no data available. "Null" values may and usually does, mix the model - I removed all null values from all data tables. - If same command is used to all the cells in table, then there is no need to repeat that command in all the cells - just one command that refers to the table is enough (just like in durations node) - I used local variable (command var:=x) in number of nodes - this way number of nodes can be reduced
I think your proposal for the cooperation sounds good. We (me or Olli) could do the most difficult programming part and you could concentrate on structure and data. How do we continue from this? The best way would be to meet and work together. If you would have data and model structure ready, I believe in one or two days we could build quite a good DALY model for Analytica. But what would be good opportunity for this work? During some intarese meeting/workshop?
Or then we could continue by emails. Could you describe a bit more detail on how detailed the model would/should be? For example, if we include life-table calculations, would it be for every year, every five years or so?
Anne Knol wrote: > Hi Marko, > Summer in Holland is not as great as in Finland I guess! We have been > having rain and thunderstorms for over a week now, and it's not about to > change...
> Anyway, about the DALYs. I have started making a model, but never finished > it. There is no actual data in the model, for example. I'm afraid my > Analytica skills are not yet completely up to the task. I have talked to > Olli about this when we had the workweek in Kuopio. We were planning to > discuss his and my DALY/Analytica plans there in order to find a way to > wrok together. Unfortunately, Olli was to busy at that time and we have not > been able to talk.
> I hereby send you the model, which is, as I said, very rudimentary. Any > comments and additions would be very useful. The MNP is trying to include > life table analysis and more detailed data in the model, but I don't think > they are very far yet. I would really promote cooperation on this issue. I > could provide some DALY 'knowledge' and perhaps data and could help > thinking about the structure of the model and what it should be able to do. > It would be great if you could do the more complicated part of the > programming and perhapd cobine the models? > Let me know your thoughts,
> Kind wishes, > Anne > > (See attached file: DALYs6.ANA) >
> > > Hi Anne, > > How is summer in Holland? Last week we had +28 here in Finland which is > quite high temperature overall and especially in early June...
> I heard from Jouni (who heard it from somewhere else) that you have done > DALY calculations with Analytica? Would it be possible to see the > Analytica model? My colleague Olli Leino has been developing a DALY > model for our group so this way we could compare the models.
> Cheers, > Marko