fit223_ls {lazy.irt} | R Documentation |
Conversion of 3PLM Items to 2PLM Items
fit223_ls( param3, wtype = 0, wmean = 0, wsd = 1, DinP = 1, npoints = 21, thmin = -3, thmax = 3, maxiter = 100, eps = 1e-06, print = 1, plot = 0, debug = 0 )
param3 |
Item Parameter Data Frame for 3PLM Items |
wtype |
= 1 to use dnorm(theta) as the weight |
wmean |
The mean of normal distribution to be used as the weight |
wsd |
The sd of normal distribution to be used as the weight |
DinP |
= 1 to include D=1.7 in logistic function |
npoints |
# of discrete points for theta |
thmin |
Minimum value of discrete thata value |
thmax |
Maximum value of discrete thata value |
maxiter |
Maximum # of GN iterations |
eps |
Convergence criterion for the relative improvement of rmse |
print |
= 1 to print result |
plot |
= 1 to plot result |
debug |
= 1 to print intemediate result |
This function minimizes
rss=sum( w*( vec(icrf_3(theta)) - vec(icrf_2(theta)) )^2 )
with respec to the 2PLM item parameters,
where icrf_3(theta)
is the icrf of input 3PLM items and
icrf_2(theta)
is the icrf of fitted 2PLM items,
and w
is the weight vector normal or 1.
Unlike fitP2G_ls and fitG2P_ls, the icrf of the 0-th category is not used.
Weighted Gauss-Newton method is used for the minimization.
A list of:
param3: Input 3PLM item parameter data frame (subset, type="P")
param2: 2PLM Item Parameter Data Frame (type="G")
rmse: Vector of sqrt(rss/length(theta)) for each item.
grad: Gradient matrix
wtype, wmean, wsd
param2 <- fit223_ls( paramS2, plot=1, print=1 ) param21 <- fit223_ls( paramS2, plot=1, print=1, wtype=1 )