比较 R 中的许多相关性

Question

我有数百个种群(不同的物种、位置等(随时间变化的相关性。如何统计分析所有这些相关性(针对它们各自的组(？

我有 600+ 相关性中每个相关性的 p 值、置信区间、相关系数，但我如何在组内比较这些(物种、位置等(。

我试图对与种群数量(百分比(和年份相关的数据进行大的偏相关，当在家庭内相关时，物种和位置(变成虚拟变量(作为控制因素，当在物种内相关时，与位置(作为虚拟变量(作为受控变量，但不确定这是否正确。

以下是一些示例数据，但这是一个与实际数据问题相反的更后勤问题：

Location    Species                 Year    Section Total   Percent Family
Chiswell A  Black-legged Kittiwake  1976    Central 310 16.78397401 Gull
Chiswell A  Black-legged Kittiwake  1986    Central 982 53.16729832 Gull
Chiswell A  Black-legged Kittiwake  1989    Central 555 30.04872767 Gull
Chiswell B  Common Murre            1976    Central 150 20.6043956  Auk
Chiswell B  Common Murre            1986    Central 385 52.88461538 Auk
Chiswell B  Common Murre            1991    Central 193 26.51098901 Auk
Natoa Is    Black-legged Kittiwake  1976    Central 70  12.77372263 Gull
Natoa Is    Black-legged Kittiwake  1986    Central 168 30.65693431 Gull
Natoa Is    Black-legged Kittiwake  1989    Central 310 56.56934307 Gull
Wooded Is   Black-legged Kittiwake  1972    Central 1560    13.35159192 Gull
Wooded Is   Pelagic Cormorant       1972    Central 144 45.56962025 Cormorant
Wooded Is   Black-legged Kittiwake  1976    Central 2350    20.11297501 Gull
Wooded Is   Glaucous-winged Gull    1976    Central 150 13.83763838 Gull
Wooded Is   Horned Puffin           1976    Central 30  33.33333333 Auk
Wooded Is   Pelagic Cormorant       1976    Central 44  13.92405063 Cormorant
Wooded Is   Pigeon Guillemot        1976    Central 100 37.59398496 Auk
Wooded Is   Red-faced Cormorant     1976    Central 4   66.66666667 Cormorant
Wooded Is   Tufted Puffin           1976    Central 9600    98.21976673 Auk
Wooded Is   Black-legged Kittiwake  2000    Central 5587    47.81752824 Gull
Wooded Is   Glaucous-winged Gull    2000    Central 472 43.54243542 Gull
Wooded Is   Horned Puffin           2000    Central 30  33.33333333 Auk
Wooded Is   Pelagic Cormorant       2000    Central 64  20.25316456 Cormorant
Wooded Is   Pigeon Guillemot        2000    Central 83  31.20300752 Auk
Wooded Is   Red-faced Cormorant     2000    Central 1   16.66666667 Cormorant
Wooded Is   Tufted Puffin           2000    Central 87  0.890116636 Auk
Wooded Is   Black-legged Kittiwake  2001    Central 2187    18.71790483 Gull
Wooded Is   Glaucous-winged Gull    2001    Central 462 42.6199262  Gull
Wooded Is   Horned Puffin           2001    Central 30  33.33333333 Auk
Wooded Is   Pelagic Cormorant       2001    Central 64  20.25316456 Cormorant
Wooded Is   Pigeon Guillemot        2001    Central 83  31.20300752 Auk
Wooded Is   Red-faced Cormorant     2001    Central 1   16.66666667 Cormorant
Wooded Is   Tufted Puffin           2001    Central 87  0.890116636 Auk

任何建议不胜感激，谢谢！

Answer 1

这超出了堆栈溢出问题的范围，可能应该发布在更面向统计的论坛中。

但是，我会查看vegan包，并专门研究规范对应分析或主成分分析。 像这样的排序方法允许您降低维数并查看许多不同分组变量之间的显着/有用的相关性。

純素主義者

更新

将时间序列作为模型中的变量之一，确定是否存在与随时间变化的重复集合相关的效应在统计上是正确的。 多向方差分析在确定变量之间的相关性方面可能比在排序方法中查找点组更有用。