Description Usage Arguments Details Value Author(s) References Examples
Bootstrap estimates of confidence regions for the bounds on model coefficients.
1 2 3 |
object |
Fitted model of class |
parm |
Names or indices of the coefficients for which to estimate confidence sets. If not specified, uses all coefficients. |
level |
Confidence level. |
type |
Whether to return the DU confidence set or the CC confidence collections. See Details for explanation. |
... |
Not used. |
Two types of confidence region are available, both defined in Beresteanu and
Molinari (2008, Section 2). The default is "DU"
, the confidence
region based on the directed Hausdorff distance. We would fail to reject
the null hypothesis that B is a subset of the population
identification region for the coefficient β_j if and only if
B \subseteq DU(β_j). The other option is "CC"
, the
confidence collection based on the undirected Hausdorff distance. We would
fail to reject the null hypothesis that [b_L, b_U] equals the
population identification region for β_j if and only if b_L
\in CC_L(β_j) and b_U \in CC_U(β_j).
The "DU"
confidence region corresponds to
interval_hypothesis
with type = "subset"
. The
"CC"
confidence region corresponds to
interval_hypothesis
with type = "equal"
.
Implements largely the same functionality as CI1D
in Beresteanu et
al.'s (2010) Stata program.
A matrix containing the range of the confidence region for each term
requested. When type== "CC"
, there are two entries per term
(range of the confidence region for the lower and upper bound).
Brenton Kenkel
Arie Beresteanu and Francesca Molinari. 2008. "Asymptotic Properties for a Class of Partially Identified Models." Econometrica 76 (4): 763–814.
Arie Beresteanu, Francesca Molinari and Darcy Steeg Morris. 2010. "Asymptotics for Partially Identified Models in Stata." https://molinari.economics.cornell.edu/programs.html
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | ## Simulate data
set.seed(18)
x1 <- rnorm(50)
x2 <- rnorm(50)
y <- 1 - x1 + x2 + rnorm(50)
yl <- floor(y)
yu <- ceiling(y)
## Fit model
fit <- coefbounds(yl + yu ~ x1 + x2, boot = 100)
## Calculate DU confidence region for coefficient on x1
confint(fit, parm = "x1")
## Calculate CC confidence collections for all coefficients
confint(fit, type = "CC")
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.