NLIV.alt {pdynmc}

Nonlinear Instrumental Variables Estimator - t-Version (NLIV.alt).

Description

NLIV.alt Computes closed form solution for lag parameter of linear dynamic panel data model based on instrumental variables (IV) estimator employing alternative formulation of nonlinear moment conditions.

Usage

NLIV.alt(dat, varname.i, varname.t, varname.y)

Arguments

dat

A dataset.

varname.i

The name of the cross-section identifier.

varname.t

The name of the time-series identifier.

varname.y

A character string denoting the name of the dependent variable in the dataset.

Details

The function estimates a linear dynamic panel data model of the form

y_{i,t} = y_{i,t-1} \rho_1 + a_i + \varepsilon_{i,t}

where y_{i,t-1} is the lagged dependent variable, \rho_1 is the lag parameter, a_i is an unobserved individual specific effect, and \varepsilon_{i,t} is an idiosyncratic remainder component. The model structure accounts for unobserved individual specific heterogeneity and dynamics. Note that more general lag structures and further covariates are beyond the scope of the current implementation in pdynmc.

The nonlinear IV estimator employs an alternative formulation of the nonlinear moment conditions of Ahn and Schmidt (1995). More details on the implementation and the properties of the estimator are provided in Fritsch et al. (2024).

Value

An object of class 'numeric' that contains the coefficient estimate for the lag parameter according to the two roots of the quadratic equation.

Author(s)

Joachim Schnurbus, Markus Fritsch

References

Ahn SC, Schmidt P (1995). “Efficient estimation of models for dynamic panel data.” Journal of Econometrics, 68(1), 5–27. doi:10.1016/0304-4076(94)01641-C.

Fritsch M, Pua AAY, Schnurbus J (2024). “Properties of an estimator for linear dynamic panel data models based on nonlinear moment conditions.” Working Paper.

Examples

## Load data
data(cigDemand, package = "pdynmc")
dat <- cigDemand

## Code example
m1 <- NLIV.alt(dat = dat, varname.i = "state", varname.t = "year", varname.y = "packpc")