statsmodels.genmod.families.links.probit class statsmodels.genmod.families.links.probit(dbn=) [source] The probit (standard normal CDF) transform Notes g(p) = scipy.stats.norm.ppf(p) probit is an alias of CDFLink. Methods deriv(p) Derivative of CDF link deriv2(p) Second derivative of the link function g??(p) inverse(z) The inverse of the CDF link inverse_deriv(z) Derivative of the inverse of the CDF transformation link function

statsmodels.genmod.families.links.Power class statsmodels.genmod.families.links.Power(power=1.0) [source] The power transform Parameters: power : float The exponent of the power transform Notes Aliases of Power: inverse = Power(power=-1) sqrt = Power(power=.5) inverse_squared = Power(power=-2.) identity = Power(power=1.) Methods deriv(p) Derivative of the power transform deriv2(p) Second derivative of the link function g??(p) inverse(z) Inverse of the power transform link function in

statsmodels.genmod.families.links.NegativeBinomial class statsmodels.genmod.families.links.NegativeBinomial(alpha=1.0) [source] The negative binomial link function Parameters: alpha : float, optional Alpha is the ancillary parameter of the Negative Binomial link function. It is assumed to be nonstochastic. The default value is 1. Permissible values are usually assumed to be in (.01, 2). Methods deriv(p) Derivative of the negative binomial transform inverse(z) Inverse of the negative bi

statsmodels.genmod.families.links.nbinom class statsmodels.genmod.families.links.nbinom(alpha=1.0) [source] The negative binomial link function. Notes g(p) = log(p/(p + 1/alpha)) nbinom is an alias of NegativeBinomial. nbinom = NegativeBinomial(alpha=1.) Methods deriv(p) Derivative of the negative binomial transform inverse(z) Inverse of the negative binomial transform inverse_deriv(z) Derivative of the inverse of the negative binomial transform

statsmodels.genmod.families.links.Logit class statsmodels.genmod.families.links.Logit [source] The logit transform Notes call and derivative use a private method _clean to make trim p by machine epsilon so that p is in (0,1) Alias of Logit: logit = Logit() Methods deriv(p) Derivative of the logit transform deriv2(p) Second derivative of the link function g??(p) inverse(z) Inverse of the logit transform inverse_deriv(z) Derivative of the inverse of the logit transform

statsmodels.genmod.families.links.Log class statsmodels.genmod.families.links.Log [source] The log transform Notes call and derivative call a private method _clean to trim the data by machine epsilon so that p is in (0,1). log is an alias of Log. Methods deriv(p) Derivative of log transform link function deriv2(p) Second derivative of the link function g??(p) inverse(z) Inverse of log transform link function inverse_deriv(z) Derivative of the inverse of the log transform link function

statsmodels.genmod.families.links.Link class statsmodels.genmod.families.links.Link [source] A generic link function for one-parameter exponential family. Link does nothing, but lays out the methods expected of any subclass. Methods deriv(p) Derivative of the link function g?(p). deriv2(p) Second derivative of the link function g??(p) inverse(z) Inverse of the link function. inverse_deriv(z) Derivative of the inverse link function g^(-1)(z).

statsmodels.genmod.families.links.inverse_squared class statsmodels.genmod.families.links.inverse_squared [source] The inverse squared transform Notes g(p) = 1/(p**2) Alias of statsmodels.family.links.Power(power=2.) Methods deriv(p) Derivative of the power transform deriv2(p) Second derivative of the link function g??(p) inverse(z) Inverse of the power transform link function inverse_deriv(z) Derivative of the inverse of the power transform

statsmodels.genmod.families.links.inverse_power class statsmodels.genmod.families.links.inverse_power [source] The inverse transform Notes g(p) = 1/p Alias of statsmodels.family.links.Power(power=-1.) Methods deriv(p) Derivative of the power transform deriv2(p) Second derivative of the link function g??(p) inverse(z) Inverse of the power transform link function inverse_deriv(z) Derivative of the inverse of the power transform

statsmodels.genmod.families.links.identity class statsmodels.genmod.families.links.identity [source] The identity transform Notes g(p) = p Alias of statsmodels.family.links.Power(power=1.) Methods deriv(p) Derivative of the power transform deriv2(p) Second derivative of the link function g??(p) inverse(z) Inverse of the power transform link function inverse_deriv(z) Derivative of the inverse of the power transform