Hello,
I am studying the nutrition status of households using a panel data survey of South-African Households. I would like to study the changes in the nutritional status of households between years.
More precisely, I divided the households into 4 different categories: Double-burden household (1), obese household (2), underweight household (3), and normal household (4).
My aim is to know the probability of a household to remain in a category (for example for an obese household to remain obese from one year to the other) or to move to another category (for example for a normal household to move to the obese category).
My regression would then be :
Nutritional status of a household (t) = nutritional status of a household (t-1) + Xt (some control variables: socioeconomic status, the gender of the household head ...)
The nutritional status of the household in t-1 (the interest variable) will then be a lagged variable.
I'm also thinking about using some household fixed effects to take into account the heterogeneity between households related to genetics.
My question concerns the model to use, do you have any advice on an econometrics model to use? Is there a model more suited than another?
I thought about using a multinomial logit regression, but maybe there are other models that are better suited for this kind of equation. Indeed, I think that the fact that there is a lagged variable may need another model?
Thank you.
Adrien
Article Explaining Fixed Effects: Random Effects Modeling of Time-Se...
https://gul.gu.se/public/pp/public_file_archive/archive.html?publishedItemId=30026152&courseId=70156&fileId=30025891
https://www.univ-orleans.fr/deg/masters/ESA/CH/Geneve_Chapitre2.pdf
https://www.cemmap.ac.uk/uploads/cemmap/programmes/Background%20reading%20May%202016.pdf
http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=9452105&fulltextType=RA&fileId=S2049847014000077
I think you are looking for panel ordinal logistic regression.
Its stata command is xtologit. You can try.
https://www.federalreserve.gov/pubs/feds/1997/199703/199703pap.pdf
Ali Hameed Yousif Thank you for this file who helped me. Do you know if there exist a special command that can be implementing on stata to do this kind of model ? Or if there is a paper that deals with it ? –
Seyyed Amir Yasin Ahmadi I believe I'm not using and ordered variable so xtologit might not be a good solution ?
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression