[R-sig-Geo] Spatial and multilevel model with kriging/interpolation in R
Thierry.ONKELINX at inbo.be
Thu Sep 25 09:27:03 CEST 2014
Have a look at the INLA package (www.r-inla.org)
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest
team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
+ 32 2 525 02 51
+ 32 54 43 61 85
Thierry.Onkelinx at inbo.be
To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of.
~ Sir Ronald Aylmer Fisher
The plural of anecdote is not data.
~ Roger Brinner
The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey
Van: r-sig-geo-bounces at r-project.org [mailto:r-sig-geo-bounces at r-project.org] Namens Justice Moses K. Aheto
Verzonden: donderdag 25 september 2014 4:25
Aan: r-sig-geo at r-project.org
Onderwerp: [R-sig-Geo] Spatial and multilevel model with kriging/interpolation in R
Please, I wish to analyse a spatial data in R through multilevel approach with my main primary objective been to interpolate for unsampled locations in my study region. Children in my data set are nested within households in the study locations and my multilevel model (without spatial) showed significant household random effects hence my choice to employ spatial analysis with multilevel approach.
The need to include household random effects in my spatial model makes it a bit difficult for me to implement in R unlike the standard geostatical analysis.
I have 'SpatialPointsDataFrame' containing my geographical coordinates (longitude and latitude) as well as my response and covariates.
The spatial mixed effects model I wish to fit and interpolate is: Yij(t) = Xij(t)β +hj+S(t)+Ɛij (1)
i=individual child, j=household, X(t)= spatial referenced non-random covariates, S(t)= spatially correlated stationary Gaussian process.
Ɛij =nugget effect/measurement error, Yij(t) = response of child i in household j at location t and is a continuous variable, hj =household level random effects and β=regression coefficients (spatial trend parameter).
Specifically, S(t)~N(0,σ2H11(ɸ) ), where σ2 is the variance (partial sill), H11(ɸ) is the correlation matrix based on valid correlation function h(u; ɸ), where u is the distance between locations and ɸ is the correlation parameter (range).
hj~N(0, σ2h), where σ2h is the household level variance Ɛij~N(0,τ2), where τ2 is the nugget effect/measurement error.
I am trying to achieve the above task through geostatistical analysis but other methods which can be implemented in R are also welcomed.
Please, could somebody help me with some papers in the literature, existing packages in R which are related to my problem as well as providing me with R codes to implement this assuming someone has already done this kind of multilevel spatial regression and interpolation in R or other packages.
Many thanks for your help in advance.
Justice Moses K. Aheto
PhD Candidate in Medicine (United Kingdom) MSc Medical Statistics (United Kingdom) BSc Statistics (Ghana) HND Statistics (Ghana)
Chief Executive Officer
Statistics and Analytics Consultancy Services Ltd.
[[alternative HTML version deleted]]
R-sig-Geo mailing list
R-sig-Geo at r-project.org
* * * * * * * * * * * * * D I S C L A I M E R * * * * * * * * * * * * *
Dit bericht en eventuele bijlagen geven enkel de visie van de schrijver weer en binden het INBO onder geen enkel beding, zolang dit bericht niet bevestigd is door een geldig ondertekend document.
The views expressed in this message and any annex are purely those of the writer and may not be regarded as stating an official position of INBO, as long as the message is not confirmed by a duly signed document.
More information about the R-sig-Geo