.- help for ^rxrsimu^ STB-28: sg45 .- Generalized ridge regression: simulated losses ---------------------------------------------- ^matrix^ ^rxsigma^ ^=^ ^(^#^)^ ^matrix^ ^rxgamma^ ^=^ ^(^#,...,#^)^ ^rxrsimu^ depvar varlist [^if^ exp] [^in^ range] [^,^ ^m^steps^(^#^)^ ^q^shape^(^#^)^ ^r^escale^(^#^)^ ^s^tart^(^#^)^ ^t^ol^(^#^)^ ] To reset problem-size limits, see ^help matsize^. Description ----------- ^rxrsimu^ simulates normal-theory response values (in a variable named ^ysim^) when the true sigma (standard deviation of errors) and the true gammas (uncorrelated components) are assumed known values and then calculates the true losses associated with generalized ridge shrinkage. Options ------- ^msteps(^#^)^ specifies the number of steps per unit increase in ^MCAL^, the multi- collinearity allowance parameter; the default value is 4. The total number of steps along the generalized shrinkage path from the least squares solu- tion (^MCAL^=0) to all zero coefficients (^MCAL^=p) will thus be 1+(msteps*p). ^qshape(^#^)^ controls the shape (or curvature) of the generalized shrinkage path through likelihood space; the default value is 0, which yields Hoerl- Kennard "ordinary" ridge regression. ^qshape^=1 yields uniform shrinkage, and all ^qshape^ values between -5 and 5 are allowed. ^rescale(^#^)^ controls the scaling of the response variable and all p predictor variables in the ^varlist^. The default value is ^rescale^=1 to scale all centered variables to have sample variance 1. ^rescale^=0 should be specified to retain the original scalings. ^start(^#^)^ controls Stata's uniform() random number seed value, and the ^rxrsimu^ default value is 12345. If you make repeated ^rxrsimu^ runs without changing this seed, you will get the same pseudo-random values each time. When you do change ^start^, make it large, positive and odd. ^tol(^#^)^ specifies search convergence criterion and defaults to 0.01. Example ------- . ^matrix rxsigma = (.5)^ ...Specify true std.dev. of error terms . ^matrix rxgamma = (-.5,.1,-.1,-.4)^ ...Specify true values of uncorrelated components . ^rxrsimu mpg cylnds cubins hpower weight, q(-1)^ ...Simulate responses and calculate true shrinkage losses Author ------ Robert L. Obenchain Eli Lilly and Company Also see -------- STB: STB-28 sg45 On-line: ^help^ for @rxrrisk@ and @rxridge@