c***z
发帖数: 6348 | 1 Hi all,
I am looking at a project related to geospatial data. The sample sizes are
small and samples are highly correlated.
Can anyone give some suggestion on how to deal with these kind of data?
Thanks a lot! |
a***g
发帖数: 2761 | 2 有个专题就叫spatial statistics
不知道你熟悉吗？ |
c***z
发帖数: 6348 | 3 Not familiar at all, can you give some introductory reference?
Thanks a lot! |
a***g
发帖数: 2761 | 4 Statistics for Spatio-Temporal Data
这本书写的不错，不过内容比较杂
Applied Spatial Data Analysis with R
这本书比较浅，但是讲了很多code，手册性质更强 |
c***z
发帖数: 6348 | |
w********e
发帖数: 944 | 6 geospatial statistics is widely used in enviornment and /or climate related
research. Geospatial data modeling can be very complex and usually
application specific. What do you exactly want to do? |
a***g
发帖数: 2761 | 7 现在做这个的bayesian很多吧
related
【在 w********e 的大作中提到】
: geospatial statistics is widely used in enviornment and /or climate related
: research. Geospatial data modeling can be very complex and usually
: application specific. What do you exactly want to do?
|
w********e
发帖数: 944 | 8 Bayesian is only one of them. Again, how complex it would be totally depends
on the purpose. For instance, you can totally model different regios
independently as the first layer, and then at the second layer, utitlize
some bayesian theory to get aggregate the results. Sounds simple, right?
【在 a***g 的大作中提到】
: 现在做这个的bayesian很多吧
:
: related
|
c***z
发帖数: 6348 | 9 Thanks a lot for the input.
This project is related to oil and gas drilling, e.g. where to drill and how
to drill.
I am not a domain expert on it and this is about all the information I have.
Also, the data is geospatial, there are few data points and they are highly
correlated. |
c***z
发帖数: 6348 | 10 Thanks a lot for the breakdown. Can you give more details? Such as an
example?
depends
【在 w********e 的大作中提到】
: Bayesian is only one of them. Again, how complex it would be totally depends
: on the purpose. For instance, you can totally model different regios
: independently as the first layer, and then at the second layer, utitlize
: some bayesian theory to get aggregate the results. Sounds simple, right?
|
w********e
发帖数: 944 | 11 For example, you have data from multiple regions of temperature, hospital
admission counts, etc), you may need to have an idea of an estimation of
of health effect (e.g., the % increase in admission cunts along with the
increase in temperature) at the overall level. You can get an estimate from
each region, and then build a Bayesian hierarchical model to get the overall
number from the regional numbers.
This is only an example of how to handle geo-spatial data.
BTW, thanks for BAO ZI. You don't need to do that. Appreciate it though.
【在 c***z 的大作中提到】
: Thanks a lot for the breakdown. Can you give more details? Such as an
: example?
:
: depends
|
c***z
发帖数: 6348 | 12 Thanks so much! Really appreciated! |
