SDPDm {SDPDmod} | R Documentation |
Spatial dynamic panel data lag model with fixed effects maximum likelihood estimation.
Description
This function estimates spatial panel model with fixed effects for static or dynamic model. It includes the transformation approach suggested by Yu et al (2008) and Lee and Yu (2010).
Usage
SDPDm(
formula,
data,
W,
index,
model = "sar",
effect = "individual",
ldet = NULL,
lndetspec = list(p = NULL, m = NULL, sd = NULL),
dynamic = FALSE,
tlaginfo = list(ind = NULL, tl = TRUE, stl = TRUE),
LYtrans = FALSE,
incr = NULL,
rintrv = TRUE,
demn = FALSE,
DIRtrans = FALSE
)
Arguments
formula |
a symbolic description for the (static) model to be estimated, not including the dynamic component |
data |
a data.frame |
W |
spatial weights matrix |
index |
the indexes (Names of the variables for the spatial and time component. The spatial is first and the time second.) |
model |
a models to be calculated, c("sar","sdm"), default = "sar" |
effect |
type of fixed effects, c("none","individual","time","twoways"), default ="individual" |
ldet |
type of computation of log-determinant, c("full","mc"). Default "full" for smaller problems, "mc" for large problems. |
lndetspec |
specifications for the calculation of the log-determinant for mcmc calculation. Default list(p=NULL,m=NULL,sd=NULL), if the number of spatial units is >1000 then list(p=30,m=30,sd=12345) |
dynamic |
logical, if TRUE time lag of the dependent variable is included. Default = FALSE |
tlaginfo |
specification for the time lag, default = list(ind=NULL,tl=FALSE,stl=FALSE), see details |
LYtrans |
logical, default FALSE. If the Lee-Yu transformation should be used for bias correction |
incr |
increment for vector of values for rho |
rintrv |
logical, default TRUE, calculates eigenvalues of W. If FALSE, the interval for rho is (-1,1) |
demn |
logical, if Lee-Yu transformation for demeaning of the variables to remove fixed effects is performed (only used in static models). Default FALSE |
DIRtrans |
logical, if direct transformation of variables should be used. Default, FALSE (only used in dynamic models with "twoways" effects) |
Details
Based on MatLab functions sar_jihai.m, sar_jihai_time.m and sar_panel_FE.m
In tlaginfo = list(ind = NULL, tl = TRUE, stl = TRUE):
ind i-th column in data which represents the time lag, if not specified then the lag from the dependent variable is created and the panel is reduced from nt to n(t-1)
tl logical, default TRUE. If TRUE y_{t-1}
(the lagged dependent variable in time is included)
stl logical, default TRUE. If TRUE Wy_{t-1}
(the lagged dependent variable in space and time is included)
Value
An object of class "SDPDm"
coefficients |
coefficients estimate of the model parameters (coefficients1 for dynamic model) |
rho |
spatial coefficient |
sige |
residuals variance |
llik |
the value of the log likelihood function |
... |
Author(s)
Rozeta Simonovska
References
Yu, J., De Jong, R., & Lee, L. F. (2008). Quasi-maximum likelihood estimators for spatial dynamic panel data with fixed effects when both n and T are large. Journal of Econometrics, 146(1), 118-134.
Lee, L. F., & Yu, J. (2010). Estimation of spatial autoregressive panel data models with fixed effects. Journal of Econometrics, 154(2), 165-185.
Lee, L. F., & Yu, J. (2010). A spatial dynamic panel data model with both time and individual fixed effects. Econometric Theory, 564-597.
See Also
vignette("spatial_model", package = "SDPDmod")
Examples
library("SDPDmod")
data(Produc, package = "plm")
data(usaww, package = "splm")
form1 <- log(gsp) ~ log(pcap) + log(pc) + log(emp) + unemp
mod1 <- SDPDm(formula = form1, data = Produc, W = usaww, index = c("state","year"),
model = "sar", effect = "individual", LYtrans = TRUE)
summary(mod1)
imp1 <- impactsSDPDm(mod1)
summary(imp1)
mod2 <- SDPDm(formula = form1, data = Produc, W = usaww, index = c("state","year"),
model = "sdm", effect = "twoways", LYtrans = TRUE,
dynamic = TRUE, tlaginfo=list(ind = NULL, tl = TRUE, stl = TRUE))
summary(mod2)