I found a paper on the effects of immigration on house prices, it uses fixed effects as a model. When looking at the results I found that the models use region fixed effects while some use year fixed effects but some use something called region-year fixed effects. what is the difference? and how can this be done in r?
I have attached an image of the results (focus on the highlighted part)
To clarify my question, my concern is that how can the model be region and year fixed effects and be region-year fixed effects at the same time.
I think you’re correct to question that. It should be either A) Year FE and Region FE, or B) Region-Year FE. Including all three, as in columns 3 and 4, should result in some being dropped to prevent collinearity issues.
Region-Year FE is equivalent to including an in indicator variable equal to 1 if an observation is in a particular region in a given year and 0 otherwise. Once all of these indicators are included, there is no additional unique information that could be contributed by Year FE or Region FE. Some software packages would still estimate the regression, but they’ll note the Year FE and Region FE are dropped. It could be that whoever prepared the table manually entered the YES/NO text and wrote it incorrectly.
Year FE is the effect of year. Region FE is the effect of region. I guess Year-Region FE is an interaction term between region and year. In other words, the effect of year might differ among regions. The model is probably something such as : Y = b1*Year + b2*Region + b3*Year*Region + bi*X.
Dear Mourad Alkalza , I also think you can assume that Year-Region fixed effect is an interaction term between region and year.
Steve can you please explain or suggest the best option? What points we need to consider to make results meaningful
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis