SummariesModel: R6 class for fitting and predicting model P(sA|sW) under g.star or g.0
Description
This R6 class Class for defining, fitting and predicting the probability model
P(sA|sW) under g_star or under g_0 for summary measures
(sW,sA). Defines and manages the factorization of the multivariate conditional
probability model P(sA=sa|...) into univariate regression models
sA[j] ~ sA[j-1] + ... + sA[1] + sW. The class self$new method automatically
figures out the correct joint probability factorization into univariate conditional
probabilities based on name ordering provided by (sA_nms, sW_nms).
When the outcome variable sA[j] is binary, this class will automatically call
a new instance of BinOutModel class.
Provide self$fit() function argument data as a DatNet.sWsA class object.
This data will be used for fitting the model P(sA|sW).
Provide self$fit() function argument newdata (also as DatNet.sWsA class) for predictions of the type
P(sA=1|sW=sw), where sw values are coming from newdata object.
Finally, provide self$predictAeqa function newdata argument
(also DatNet.sWsA class object) for getting the likelihood predictions P(sA=sa|sW=sw), where
both, sa and sw values are coming from newdata object.
Methods
new(reg, ...)- ...
length- ...
getPsAsW.models- ...
getcumprodAeqa- ...
copy.fit(SummariesModel)- ...
fit(data)- ...
predict(newdata)- ...
predictAeqa(newdata, ...)- ...
Active Bindings
wipe.alldat- ...
Details
n_regs - .
parfit_allowed - .