schmittbin -- Compute regression coefficients by Scmitt's method.
The schmittbin task calculates the binned two-dimensional Kaplan-Meier distribution and associated linear regression coefficients derived by Schmitt (1985). Schmitt's binned linear regression can treat mixed censoring including censoring in the independent variable, but can have only one independent variable. This task requires only that the censoring distribution about the fitted line is random.
The principal outputs of this task are the estimates of the intercept and slope of the regression line and the standard deviations of these estimates.
Schmitt's binned regression has a number of drawbacks discussed by Sadler et al. (1989), including slow or failed convergence to the two-dimensional Kaplan-Meier distribution, arbitrary choice of bin size and origin, and problems if either too many or too few bins are selected. We suggest that Schmitt's regression be reserved for problems with censoring present in both variables.
- input [string]
- Input file(s); this can be a list of files. Following each file name is a list of column names in brackets. Thes column names specify which columns in the file contain the information used by this task. The brackets MUST contain three names in the following format: [censor_indicator, independent_var, dependent_var]. The censor indicator specifies the censorship of the data point. The different kinds of censorship are explained in the censor help file. The second name in the brackets specifies the column containing the independent variable. The third name specifies the column containing the dependent variable. If the input file is an STSDAS table, the names in brackets are the table column names. If the input file is a text file, the names in brackets are the column numbers. A title string will be printed if the input file is a table containing the header parameter TITLE.
- (nxbins = 10) [integer, min=1, max=40]
- Number of bins in independent variable.
This should be chosen carefully because the larger this number, the longer the task will take to run.
- (nybins = 10) [integer, min=1, max=40]
- Number of bins in dependent variable.
This should be chosen carefully because the larger this number, the longer the task takes to run.
- (xsize = 0.) [real]
- Size of bins in the independent variable.
If the value of this parameter or ysize is zero, the program will choose the bin sizes and origins (xsize, ysize, xorigin, and yorigin) based on the range of the data and the number of bins.
- (ysize = 0.) [real]
- Size of bins in dependent variable. If the value of this parameter or xsize is zero, the program will choose the bin sizes and origins.
- (xorigin = 0.) [real]
- Origin of bins in independent variable.
- (yorigin = 0.) [real]
- Origin of bins in dependent variable.
- (tolerance = 1.0000000000000E-5) [real, min=0.]
- Tolerance for regression fit. If the sum of the squares of the differences between two successive estimates of the probability function is greater than the tolerance, the iteration stops.
- (niter = 50) [integer, min=1]
- Maximum number of iterations. The regression is computed from the probability density function, which is computed by an iterative alogorithm. The computation stops when the maximum number of iterations is performed even if convergence has not been achieved.
- (nboot = 200) [integer, min=0]
- Number of iterations of the bootstrap method used in computing the error in the estimates of the regression coefficients. If the value of this parameter is set to zero, the bootstrap regression coefficient errors will not be computed. The running time of this task will increase proportionately with the value of this parameter.
- (verbose = no) [boolean]
- Print a detailed listing?
A detailed display includes the distribution of bins, sizes and number of points in each.
1. Apply Schmitt's binned method to the data in the table "kriss.tab", using the columns "Censor" for the censoring indicator, "LogL1mu" for the first independent variable column, "LogL2500A" for the second independent variable column, and "LogL2keV" for the dependent variable column. Then use the file "iraslum.dat" (text file), columns 1, 2, and 3 (censor, independent, dependent). Several files may be processed sequentially. The following example will compute the Schmitt's binned regression for the two files indicated:
cl> schmittbin kriss.tab[Censor,LogL1mu,LogL2500A,LogL2keV], \ >>> iraslum.dat[1,2,3]
Type "help statistics option=sys" for a higher-level description of the statistics package.