CreateInitializeMatrix: Create Possible Path Matrix
In PathSelectMP: Backwards Variable Selection for Paths using M Plus

Description Usage Arguments Details Value Note Author(s) References Examples

creates a dataframe with 1's and 0s' where a 1 represents inclusion in the path and 0 represents exlusion

1	CreateInitializeMatrix(InitialData, WhichCat, empty = FALSE)

`InitialData`	the dataframe with the initial data
`WhichCat`	a vector of 0's and 1's length of number of columns of `InitialData` where 1 is a categorical variable and 0 is a continuous variable
`empty`	if `empty` is TRUE then a dataframe of all 0's is returned to allow user to individually assign 1's for path inclusion more easily

the rows represent dependent variables in the path and the columns represent independent variables in the paths with a 1 representing in M Plus terminology row variable ON column variable

returns matrix required by Initialize for all initial paths to run the default is an upper triangular matrix. This is a square matrix and names of rows is the same as the names of columns

In this version only Categorical variables can be dependent variables and if using the default upper triangular matrix, due to the current way indirect effects are created, only approximately 8 or 9 variables can be used with reasonable computing time. The input to Initialize must be in the same format as this output so user may want to use this function with empty as FALSE and subtitute 1's for desired possible paths. Also note that row names and column names must be the same and must start with a capital letter and only contains numbers and letters in this version.

William Terry

M Plus