Description Usage Arguments See Also
various parameters that control aspects of the "OutlierDM" object
1 2 3 4 5 | odm.control(pair.cre = 1, dist.mthd = "median",
Lower = .25, Upper = .75, trans = "log2",
centering = TRUE, projection.type = "PCA", lbda = 1,
nonlin.method = "L-BFGS-B", nonlin.SS = "AsymOff",
nonlin.Frank = c(2, -8, 0, 1), ncl = 2, alpha = 0.05 )
|
pair.cre |
a scalar parameter to specify the minimum number of pairs, used in "type = pair". |
dist.mthd |
a distance parameter used in "type = diff". you can choose one of "median", "mean" or and so on. |
Lower |
a criterion for lower quantile value used to construct boxplot |
Upper |
a criterion for upper quantile value used to construct boxplot |
trans |
a parameter for a logarithm and exponential transformation. If a log 2 transformation is needed, set "trans = log2". If no transformation is needed, set "trans = FALSE". |
centering |
a logical parameter for the status of centering. If "centering = TRUE", data are centered by its column means. |
projection.type |
a parameter to determine a type of projection methods. Choose one of "naive", "pca", and "robust". |
lbda |
a criterion about lambda used for nonlinear quantile regression. |
nonlin.method |
a parameter to determine a type of methods used for nonlinear quantile regression. choose one of "L-BFGS-B" and "BFGS". Default is "L-BFGS-B". |
nonlin.SS |
a parameter to determine a type of structure used for nonlinear quantile regression. choose one of "Frank", "Self", "Asym" and "AsymOff". Default is "AsymOff", Asymptotic Regression Model with an Offset. |
nonlin.Frank |
a structure parameter used for Frank copula model. Gain c(df, delta, mu, sigma) in the Frank copula formula |
ncl |
A parameter to determine the number of cores used in parallel computing. A default value is 2. |
alpha |
A sigfinicance level of the Grubbs test. A default value is 0.05. |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.