Correlation analysis. Methods of Pearson, Spearman, Kendall and Lin — correlation • agricolae

It obtains the coefficients of correlation and p-value between all the variables of a data table. The methods to apply are Pearson, Spearman , Kendall and lin's concordance index. In case of not specifying the method, the Pearson method will be used. The results are similar to SAS.

correlation(
  x,
  y = NULL,
  method = c("pearson", "kendall", "spearman", "lin"),
  alternative = "two.sided"
)

Arguments

x	table, matrix or vector
y	table, matrix or vector
method	"pearson", "kendall", "spearman", "lin"
alternative	"two.sided", "less", "greater"

Value

The correlation matrix with its probability

Details

Parameters equal to function cor()

References

Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics. 1989; 45, 255-268.

See also

Examples


library(agricolae)
data(soil)
# example 1
analysis<-correlation(soil[,2:8],method="pearson")
analysis
#> $correlation
#>          pH    EC CaCO3    MO  CIC    P    K
#> pH     1.00  0.55  0.73 -0.33 0.26 0.14 0.59
#> EC     0.55  1.00  0.32 -0.39 0.00 0.46 0.57
#> CaCO3  0.73  0.32  1.00 -0.23 0.30 0.05 0.36
#> MO    -0.33 -0.39 -0.23  1.00 0.53 0.56 0.23
#> CIC    0.26  0.00  0.30  0.53 1.00 0.55 0.47
#> P      0.14  0.46  0.05  0.56 0.55 1.00 0.62
#> K      0.59  0.57  0.36  0.23 0.47 0.62 1.00
#> 
#> $pvalue
#>                pH        EC       CaCO3         MO        CIC          P
#> pH    1.000000000 0.0525330 0.004797027 0.27832165 0.38601765 0.65138915
#> EC    0.052532997 1.0000000 0.294159813 0.18740661 0.98782240 0.11113616
#> CaCO3 0.004797027 0.2941598 1.000000000 0.45551075 0.31509291 0.87385528
#> MO    0.278321645 0.1874066 0.455510753 1.00000000 0.06028063 0.04468908
#> CIC   0.386017651 0.9878224 0.315092905 0.06028063 1.00000000 0.05132886
#> P     0.651389152 0.1111362 0.873855279 0.04468908 0.05132886 1.00000000
#> K     0.034846847 0.0398372 0.227586184 0.44234137 0.10181733 0.02442050
#>                K
#> pH    0.03484685
#> EC    0.03983720
#> CaCO3 0.22758618
#> MO    0.44234137
#> CIC   0.10181733
#> P     0.02442050
#> K     1.00000000
#> 
#> $n.obs
#> [1] 13
#> 
# Example 2: correlation between pH, variable 2 and other elements from soil.
analysis<-with(soil,correlation(pH,soil[,3:8],method="pearson",alternative="less"))
analysis
#> $correlation
#>      EC CaCO3    MO  CIC    P    K
#> pH 0.55  0.73 -0.33 0.26 0.14 0.59
#> 
#> $pvalue
#>        EC  CaCO3     MO   CIC      P      K
#> pH 0.9737 0.9976 0.8608 0.807 0.6743 0.9826
#> 
#> $n.obs
#> [1] 13
#> 
# Example 3: correlation between pH and clay method kendall.
with(soil,correlation(pH,clay,method="kendall", alternative="two.sided"))
#> 
#> Kendall's rank correlation tau
#> 
#> data: pH and clay 
#> z-norm =  3.31896 p-value = 0.0009035325 
#> alternative hypothesis: true rho is not equal to 0
#> sample estimates:
#> tau
#>  0.7105263