Questions tagged [spatial-correlation]
The spatial-correlation tag has no usage guidance.
12
questions
1
vote
0
answers
30
views
Decreasing trend in semivariogram of nlme model residuals?
I am new to spatial statistics and am trying to fit some spatial models using the nlme package in R.
I fit three different models, using the same set of predictors ...
1
vote
0
answers
10
views
Spatial Error Model vs Conley standard errors
What is the difference (pros and cons) of using an OLS with conley standard errors and a spatial error model? Is there a recommendation for when you should use one over the other?
Thanks!
0
votes
0
answers
68
views
Metric to quantify randomness
I have an $L \times L$ matrix representing a $2D$ region. Each entry of this matrix is a real number lying in the closed interval [0,1]. I want to quantify how different is this from a similar $L \...
0
votes
0
answers
12
views
Do spatial regression models hold the assumption of linearity?
I'm running a spatial regression model (probably spatial durbin error model or spatial durbin model) and one of the explanatory continuous variables has a lot of (true) zeros. I'm worried that the ...
1
vote
1
answer
116
views
Expectation and Variance of Moran's I under the Null
Moran's I is a statistic used to measure spatial autocorrelation. For a set of $N$ spatial units where we get measurements $\mathbf{x} = (x_1, x_2, \cdots, x_N)^T$, and a weight matrix between the ...
0
votes
0
answers
43
views
R package for spatial regression
I want to model counts of a beetle (B) as a function of three continuous dependent variables (PC1, PC2, RL), weighted by a fourth variable (W) using either poisson or quasipoisson regression with ...
0
votes
1
answer
32
views
What means that the model is "singular"?
My question might be rather basic, theoretical.
I am running spatial and spatial-temporal bayesian models in INLA. I have areal data and a continuous response variable with spatial and temporal ...
6
votes
1
answer
140
views
References on data partitioning (cross-validation, train/val/test set construction) when data are non-IID
Consider a prediction setting in which we are interested in training a regression or classification function $f$ with inputs $X \in \mathbb{R}^k$ and target $Y$, and assessing its expected ...
0
votes
0
answers
17
views
P-Value of constant in Spaltial Lag Model is too high
I ran the OLS model in my dataset, all tests for residuals suggested that the older is not fit. When running Spatial Lag Model I got the results as following:
...
0
votes
0
answers
44
views
INLA model with spatial autocorrelation - binary response, binomial predictor
I have a binary response variable (0/1) and a predictor distributed continuously on 0-1 scale. There is significant spatial autocorrelation in data, thus I am running an INLA model to account for that....
0
votes
0
answers
29
views
Should I use random effect or spatial correlation structure?
I'm working on a vegetation analysis, where I am comparing different remote sensing indices' ability to predict field measured vegetation characteristics. Sample size is 56.
The study design:
Two ...
2
votes
0
answers
82
views
What are the types of error correlation?
In many fundamental regression models errors are assumed Independent Identically Distributed.
One of violations of this assumption is when the errors are somehow correlated. From what I know it may ...