spatialrisk {spatrisk} | R Documentation |
spatialrisk
is estimates the model described in Lewis and Berinsky (2007).
spatialrisk(y,x,issues,controls,integrationSampleSize, issuelab=sapply(1:100,function(i) sprintf("Issue controllab=sapply(1:100,function(i) sprintf("Control riskvar=FALSE, param=c(1.0, 0.0, 0.0, 0.0, 0.1,rep(0,issues),rep(0,controls)) ,...)
y |
vector of vote choices, -1 or 1. |
x |
matrix of covariate (specific format see examples). |
issues |
(integer) number of issues |
integrationSampleSize |
(integer) Number of Monte Carlo draws to use in each dim of each obs integral. |
issuelab |
Vector of issue labels. |
controllab |
Vector of control variable labels. |
riskvar |
True/False vector noting if x includes a riskvar column. |
normal |
integer, '1' generates data using normal probabilities, |
param |
Vector of start values. |
priors |
Length 6 vector containing three prior means and the three prior standard deviations for parameters alpha, gamma, and delta. |
... |
NLM parameters. |
Usual type output
Jeffrey Lewis jblewis@ucla.edu
library("spatrisk") data("pres76") res<-spatialrisk(pres76[,1],pres76[,2:dim(pres76)[2]],issues=3,controls=3, integrationSampleSize=5,riskvar=FALSE, print.level=2)