StatTools.cs

Use TMtxLogistReg component to perform ordinary or ordinal logistic regression on dataset.Suppose y takes values in k ordered categories, and let p_ij be the cumulative probability that y(i) falls in the j'th category or higher. The ordinal logistic regression model is defined as: 

logit(p_ij) = theta(j) + A_i'B , i = 1,..,length(Y), j = 1,..,k-1,  

where A_i is the i'th row of A . The number of ordinal categories k is taken to be the number of distinct values of int(y). If k is 2 (two categories) the model is ordinary logistic regression.  

How to use TMtxLogistReg component?  

1. Drop a TMtxLogistReg component on the form.
2. Define crefA, the matrix of independent variables.
3. Define crefY, grouping (categories) vector. Note: All values have to be integers.
4. Define initial estimates for crefTheta and crefB coefficients. Alternatively you can also set crefAutoInitEstimates to true and let the algorithm calculate initial estimates for Theta and B.
5. (Optionally) Set crefTolerance to define desired B, Theta estimates precision. Set crefMaxIteration to define maximum number of iterations in main calculation loop.
6. Call the crefRecalc method to trigger calculation.

Results:

1. crefB : B coefficients estimates.
2. crefTheta : Theta coefficients estimates.
3. crefTBStdErr : Theta and B coefficients estimates standard errors.

How to setup and run logistic regression? In the following example we'll perform ordinary (i.e. with two categories) logistic regression on data, taken from NCSS statistics program. Note that TMtxLogistReg.Y vector values are binary integer values.

Classes