Instrumental-variable estimation of large-T panel-data models with common factors

Abstract

In this article, we introduce the xtivdfreg command, which implements a general instrumental-variables (IV) approach for fitting panel-data models with many time-series observations, T, and unobserved common factors or interactive effects, as developed by Norkute et al. (2021, Journal of Econometrics 220: 416–446) and Cui et al. (2020a , ISER Discussion Paper 1101). The underlying idea of this approach is to project out the common factors from exogenous covariates using principal-components analysis and to run IV regression in both of two stages, using defactored covariates as instruments. The resulting two-stage IV estimator is valid for models with homogeneous or heterogeneous slope coefficients and has several advantages relative to existing popular approaches. In addition, the xtivdfreg command extends the two-stage IV approach in two major ways. First, the algorithm accommodates estimation of unbalanced panels. Second, the algorithm permits a flexible specification of instruments. We show that when one imposes zero factors, the xtivdfreg command can replicate the results of the popular Stata ivregress command. Notably, unlike ivregress, xtivdfreg permits estimation of the two-way error-components paneldata model with heterogeneous slope coefficients.

The impact of child mortality on fertility in South Africa: Do child support grants and antiretroviral treatment matter?

“…Table 4 reports the pooled OLS, RE and FE estimates of the effect of under-five mortality, CSG and ART on TFR using Driscoll and Kraay standard errors. In addition, Table 4 also reports 2SIV estimator results obtained using the xtivdfreg Stata command of Kripfganz and Sarafidis [73]. We observe from Table 4 that the regression results reported in column (1) to (3) are qualitatively and quantitatively similar to those of the pooled OLS, RE and FE models obtained…”

“…To account for potential endogeneity of some of the regressors (CSG and ART), we also provide results of the two-stage instrumental variables estimator (2SIV) for large-T panel data models, as developed by Norkute et al [ 72 ]. The 2SIV estimator uses defactored covariates as instruments and provides a flexible specification of instruments [ 73 ]. When estimating the 2SIV model, we use the mean group (MG) estimator that considers potential heterogeneity of slope coefficients.…”

“…Table 4 reports the pooled OLS, RE and FE estimates of the effect of under-five mortality, CSG and ART on TFR using Driscoll and Kraay standard errors. In addition, Table 4 also reports 2SIV estimator results obtained using the xtivdfreg Stata command of Kripfganz and Sarafidis [ 73 ].…”

Panel unit-root tests with structural breaks

“…In this section, we focus on bank balance sheet variables taken from the call reports of the Federal Deposit Insurance Corporation. This is a very popular dataset; see, for example, Kripfganz and Sarafidis (2021) and Juodis, Karavias, and Sarafidis (2021) for some recent applications. However, while stationarity of the bank balance sheet variables is frequently assumed, it is almost never tested.…”

“…The xtbunitroot command is applied to examine stationarity in four fundamental bank balance sheet variables frequently used in the banking literature: returns on assets, returns on equity, total assets, and noninterest income. Their stationarity properties affect model building and economic evaluation; see, for example, Kripfganz and Sarafidis (2021) and Delis and Karavias (2015), and in panel forecasting, see, for example, Liu, Moon, and Schorfheide (2020). We examine a sample of 500 randomly selected U.S. banks for the period 2018Q3 to 2020Q4.…”