generate_table: Generate diagnostic tables for predictors and predictands.
In leerichardson/sdsmR: Implementation of Statistical Downscaling

Description Usage Arguments Value Examples

This function is meant to replicate the screen variables section of the SDSM tool. It creates two plots, the table which shows the correlations between predictors in all individual months and annually, as well as the correlations between predictor variables in all months.

1 2	generate_table(plotname, dataframe, y = "empty", conditional = FALSE, conditional_step = 1)

`plotname`	The name you want to save the plots as. IE: if your data comes from the blogsville data-set, you might want to call this "blogsville". If you want to save the plot in a different directory, just specify this in this name: "/home/blogsville" will save the plot's in the home directory.
`dataframe`	a dataframe object which contains a column of dates, the predictand (response) variable, as well as the predictor variables.
`y`	The predictand (response) variable name from the same dataframe. This must be a character!
`conditional`	Whether you want the correlations to correspond to the conditional model. If TRUE, then the function will return correlations for the response variable. The step of the conditional model it will return is determined by the conditional_step parameter.
`conditional_step`	Determines whether the first or second step of the conditional model is returned. If 1, then this will return the correlations with the 0/1 reponse variable. If 2, then the tables will correspond to the original response variable, but only on the days in which the variable is a 1.

A correlation matrix between predictor variables and two PDF file diagnostic plots.

# Generate diagnostic tables for both temperature maximum and
# precipitation.
generate_table("blogsville_explore", blogsville, y = "tmax")
generate_table("blogsville_explore", blogsville, y = "pcrp",
     conditional = TRUE, conditional_step = 1)