我正在使用R包eDNAoccupancy来开发具有三个级别的贝叶斯层次模型，但是我正在努力寻找同时使用站点级和样本级协变量的示例（仅引用是可能的） 。
我已经按照我应该的格式格式化了数据，但我总是遇到错误。
任何帮助或指导将不胜感激！
我的检测矩阵如下：
> head(hbDetections$y)
[,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12] [,13] [,14] [,15]
S1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
S2 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0
S3 0 0 0 0 0 0 1 0 0 0 2 0 1 0 1
[,16] [,17] [,18] [,19] [,20] [,21] [,22] [,23] [,24] [,25] [,26] [,27] [,28]
S1 1 0 0 0 0 0 0 0 0 0 1 0 0
S2 0 0 0 1 0 0 0 0 0 0 0 0 0
S3 0 0 0 1 0 0 0 0 0 0 0 0 0
[,29] [,30] [,31] [,32] [,33] [,34] [,35] [,36] [,37] [,38] [,39] [,40] [,41]
S1 1 0 0 2 0 0 0 0 0 0 0 0 0
S2 0 0 1 0 0 0 0 0 0 0 0 0 0
S3 0 1 1 0 0 1 0 0 1 0 0 0 NA
[,42] [,43] [,44] [,45] [,46]
S1 NA NA NA NA NA
S2 0 0 1 0 0
S3 NA NA NA NA NA
> head(hbDetections$K)
[,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12] [,13] [,14] [,15]
S1 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
S2 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
S3 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5
[,16] [,17] [,18] [,19] [,20] [,21] [,22] [,23] [,24] [,25] [,26] [,27] [,28]
S1 5 5 5 5 5 5 5 5 5 5 5 5 5
S2 5 5 5 5 5 5 5 5 5 5 5 5 5
S3 5 5 5 5 5 5 5 5 5 5 5 5 5
[,29] [,30] [,31] [,32] [,33] [,34] [,35] [,36] [,37] [,38] [,39] [,40] [,41]
S1 5 5 5 5 5 5 5 5 5 5 5 5 5
S2 5 5 5 5 5 5 5 5 5 5 5 5 5
S3 5 5 5 5 5 5 5 5 5 5 5 5 0
[,42] [,43] [,44] [,45] [,46]
S1 0 0 0 0 0
S2 5 5 5 5 5
S3 0 0 0 0 0
我的协变量如下：
注意：大小，SI和Light站点级协变量以及NDVI和TRI样本级协变量
> head(siteAndSampleData)
Site Sample Size SI Light NDVI TRI
1 S1 1 313.28 1.76 881 6398.00 188.3559
2 S1 2 313.28 1.76 881 5881.00 191.6983
3 S1 3 313.28 1.76 881 7253.25 390.0000
4 S1 4 313.28 1.76 881 6565.50 299.3047
5 S1 5 313.28 1.76 881 6332.25 287.0660
6 S1 6 313.28 1.76 881 6641.75 250.7060
然后我用occModel运行eDNAoccupancy，如下所示：
set.seed(0157)
fit = occModel(formulaSite = ~ 1,
formulaSiteAndSample = ~ NDVI,
formulaReplicate = ~ 1,
detectionMats = hbDetections,
siteData = NULL,
siteAndSampleData = siteAndSampleData.sc,
niter = 6000,
niterInterval = 2000,
siteColName = "Site",
sampleColName = "Sample"
)
我收到以下警告消息：
> warnings()
Warning messages:
1: glm.fit: fitted probabilities numerically 0 or 1 occurred
坚持不懈地奋斗，我尝试并适应后验总结
> posteriorSummary(fit, burnin = 1000, mcError = TRUE, outputSummary=TRUE)
Bayesian estimates of model parameters
Mean 50% 2.5% 97.5%
beta..Intercept. 1.042 1.015 -0.243 2.513
alpha..Intercept. 0.591 0.501 -0.614 2.230
alpha.NDVI -0.347 -0.338 -1.457 0.898
delta..Intercept. -1.618 -1.647 -1.942 -1.156
Monte Carlo SE of Bayesian estimates
Mean 50% 2.5% 97.5%
beta..Intercept. 0.0100 0.0119 0.0216 0.0277
alpha..Intercept. 0.0687 0.0859 0.0313 0.1224
alpha.NDVI 0.0406 0.0337 0.0689 0.1268
delta..Intercept. 0.0151 0.0163 0.0089 0.0222
NULL
> psi = posteriorSummaryOfSiteOccupancy(fit, burnin = 1000)
> theta = posteriorSummaryOfSampleOccupancy(fit, burnin = 1000)
最后收到一条错误消息：
Error in quantile.default(input, prob = q, type = 1, names = FALSE) :
missing values and NaN's not allowed if 'na.rm' is FALSE
再次感谢您提出关于如何使用eDNAoccupancy格式化我的数据框和/或模型以便合并站点级协变量和样本级协变量的建议。
谢谢！