calib_oneway: Find one-way (intercept) correction for cell estimates

In kuriwaki/ccesMRPrun: Using MRP Models from Cleaned CCES and Census Data

Find one-way (intercept) correction for cell estimates

Description

For a given geography j, there may be a value pi_j which is the ground truth. The analyst has C cells with estimates of pi_c that may be biased. This function, proposed by Ghitza and Gelman and Ghitza, will find a intercept shift for all ⁠C cells to best fit the total ⁠pi'. It is the argmin of the sum of absolute values of the deviation.

Usage

calib_oneway(tgt, ests, n, search = c(-5, 5))

Arguments

tgt	The true target
ests	A vector of current estimate, by cell
n	The sample size of the estimates, by cell
search	The lower and upper endpoints of the interval to search

Value

The value of delta or the intercept that minimizes the absolute deviation in total. The value is on the logit scale. To translate to a probability, use invlogit.

Author(s)

Yair Ghitza

Source

FindDelta function at https://github.com/Catalist-LLC/unemployment/blob/master/unemployment_cps_mrp/helper_functions/GetYHat.R

Also see Evan T. R. Rosenman and Santiago Olivella, "Recalibration of Predictive Models as Approximate Probabilistic Updates" https://arxiv.org/abs/2112.06674

See Also

calib_twoway

Examples

 biased_ests <- ccesMRPrun:::invlogit(rnorm(n = 100, mean = 1, sd = 1))
 sizes <- rbinom(n = 100, size = 100, prob = 0.1)
 tru <- 0.5

 calib_oneway(tgt = tru, ests = biased_ests, n = sizes)

kuriwaki/ccesMRPrun documentation built on Sept. 24, 2024, 2:15 a.m.