Rehabilitating the Lagged Dependent Variable With Structural Equation Modeling
In: Structural equation modeling: a multidisciplinary journal, Band 30, Heft 4, S. 659-671
ISSN: 1532-8007
191 Ergebnisse
Sortierung:
In: Structural equation modeling: a multidisciplinary journal, Band 30, Heft 4, S. 659-671
ISSN: 1532-8007
In: USAEE Working Paper No. 23-595
SSRN
In: Political analysis: PA ; the official journal of the Society for Political Methodology and the Political Methodology Section of the American Political Science Association, Band 27, Heft 4, S. 605-615
ISSN: 1476-4989
Difference-in-differences is a widely used evaluation strategy that draws causal inference from observational panel data. Its causal identification relies on the assumption of parallel trends, which is scale-dependent and may be questionable in some applications. A common alternative is a regression model that adjusts for the lagged dependent variable, which rests on the assumption of ignorability conditional on past outcomes. In the context of linear models, Angrist and Pischke (2009) show that the difference-in-differences and lagged-dependent-variable regression estimates have a bracketing relationship. Namely, for a true positive effect, if ignorability is correct, then mistakenly assuming parallel trends will overestimate the effect; in contrast, if the parallel trends assumption is correct, then mistakenly assuming ignorability will underestimate the effect. We show that the same bracketing relationship holds in general nonparametric (model-free) settings. We also extend the result to semiparametric estimation based on inverse probability weighting. We provide three examples to illustrate the theoretical results with replication files in Ding and Li (2019).
Difference-in-differences is a widely used evaluation strategy that draws causal inference from observational panel data. Its causal identification relies on the assumption of parallel trends, which is scale-dependent and may be questionable in some applications. A common alternative is a regression model that adjusts for the lagged dependent variable, which rests on the assumption of ignorability conditional on past outcomes. In the context of linear models, Angrist and Pischke (2009) show that the difference-in-differences and lagged-dependent-variable regression estimates have a bracketing relationship. Namely, for a true positive effect, if ignorability is correct, then mistakenly assuming parallel trends will overestimate the effect; in contrast, if the parallel trends assumption is correct, then mistakenly assuming ignorability will underestimate the effect. We show that the same bracketing relationship holds in general nonparametric (model-free) settings. We also extend the result to semiparametric estimation based on inverse probability weighting. We provide three examples to illustrate the theoretical results with replication files in Ding and Li (2019).
BASE
In: Deutsche Bundesbank Discussion Paper No. 15/2019
SSRN
Working paper
In: Political analysis: PA ; the official journal of the Society for Political Methodology and the Political Methodology Section of the American Political Science Association, Band 14, Heft 2, S. 186-205
ISSN: 1476-4989
A lagged dependent variable in an OLS regression is often used as a means of capturing dynamic effects in political processes and as a method for ridding the model of autocorrelation. But recent work contends that the lagged dependent variable specification is too problematic for use in most situations. More specifically, if residual autocorrelation is present, the lagged dependent variable causes the coefficients for explanatory variables to be biased downward. We use a Monte Carlo analysis to assess empirically how much bias is present when a lagged dependent variable is used under a wide variety of circumstances. In our analysis, we compare the performance of the lagged dependent variable model to several other time series models. We show that while the lagged dependent variable is inappropriate in some circumstances, it remains an appropriate model for the dynamic theories often tested by applied analysts. From the analysis, we develop several practical suggestions on when and how to use lagged dependent variables on the right-hand side of a model.
In: Political analysis: official journal of the Society for Political Methodology, the Political Methodology Section of the American Political Science Association, Band 14, Heft 2, S. 186-205
ISSN: 1047-1987
In: Journalism & mass communication quarterly: JMCQ, Band 84, Heft 4, S. 695-712
ISSN: 2161-430X
This study assesses news and media campaign effects during the recovery phase of a catastrophe. With data from a panel telephone survey in New Orleans in 2006, this study tests lagged dependent variable models for safety beliefs and safety behavior in the context of Hurricane Katrina. News attention and media campaign exposure influenced safety behavior. The effects of news attention were synchronous, while those of media campaign exposure were cross-lagged. In contrast, neither news attention nor media campaign exposure influenced safety beliefs, which may be attributable to ceiling effects of the belief measure. Safety beliefs did, however, have a cross-lagged influence on safety behavior.
In: Political science research and methods: PSRM, Band 6, Heft 2, S. 393-411
ISSN: 2049-8489
Lagged dependent variables (LDVs) have been used in regression analysis to provide robust estimates of the effects of independent variables, but some research argues that using LDVs in regressions produces negatively biased coefficient estimates, even if the LDV is part of the data-generating process. I demonstrate that these concerns are easily resolved by specifying a regression model that accounts for autocorrelation in the error term. This actually implies that more LDV and lagged independent variables should be included in the specification, not fewer. Including the additional lags yields more accurate parameter estimates, which I demonstrate using the same data-generating process scholars had previously used to argue against including LDVs. I use Monte Carlo simulations to show that this specification returns much more accurate coefficient estimates for independent variables (across a wide range of parameter values) than alternatives considered in earlier research. The simulation results also indicate that improper exclusion of LDVs can lead to severe bias in coefficient estimates. While no panacea, scholars should continue to confidently include LDVs as part of a robust estimation strategy.
In: Political analysis: PA ; the official journal of the Society for Political Methodology and the Political Methodology Section of the American Political Science Association, Band 29, Heft 4, S. 561-569
ISSN: 1476-4989
AbstractDebate on the use of lagged dependent variables has a long history in political science. The latest contribution to this discussion is Wilkins (2018, Political Science Research and Methods, 6, 393–411), which advocates the use of an ADL(2,1) model when there is serial dependence in the outcome and disturbance. While this specification does offer some insurance against serially correlated disturbances, this is never the best (linear unbiased estimator) approach and should not be pursued as a general strategy. First, this strategy is only appropriate when the data-generating process (DGP) actually implies a more parsimonious model. Second, when this is not the DGP—e.g., lags of the predictors have independent effects—this strategy mischaracterizes the dynamic process. We clarify this issue and detail a Wald test that can be used to evaluate the appropriateness of the Wilkins approach. In general, we argue that researchers need to always: (i) ensure models are dynamically complete and (ii) test whether more restrictive models are appropriate.
SSRN
In: Alcohol and alcoholism: the international journal of the Medical Council on Alcoholism (MCA) and the journal of the European Society for Biomedical Research on Alcoholism (ESBRA), Band 49, Heft 6, S. 645-653
ISSN: 1464-3502
The purpose of this master's thesis is to understand the time-variation in the correlations between U.S. stock and government bond returns. In particular, an underlying objective of this empirical research is to comprehend the behavior of the correlation during financial stress periods. The model that is used to estimate these correlations is the Multivariate DCC-GARCH model as first introduced by Engle in 2001. Eighteen macro-finance factors are selected from the existing literature. Statistical significance of these factors' impact on the stock-bond correlation is tested through three proposed lagged dependent variable models. The out-of-sample forecasting performance of these models is compared against a benchmark, i.e. the random walk model. The empirical analysis is conducted at both daily and monthly frequencies for a period ranging from January 1997 until November 2014. The empirical results show that high stock volatility explains the flight-to-safety phenomenon during financial stress epochs. Moreover, the findings indicate that the short rate drives the time-variation at daily frequency and that the yield spread and stock liquidity explain the monthly dynamics in stock and government bond return correlations. This research is motivated by its crucial implications on risk management and portfolio optimization. ; Master [120] en Ingénieur de gestion, Université catholique de Louvain, 2015
BASE
The purpose of this master's thesis is to understand the time-variation in the correlations between U.S. stock and government bond returns. In particular, an underlying objective of this empirical research is to comprehend the behavior of the correlation during financial stress periods. The model that is used to estimate these correlations is the Multivariate DCC-GARCH model as first introduced by Engle in 2001. Eighteen macro-finance factors are selected from the existing literature. Statistical significance of these factors' impact on the stock-bond correlation is tested through three proposed lagged dependent variable models. The out-of-sample forecasting performance of these models is compared against a benchmark, i.e. the random walk model. The empirical analysis is conducted at both daily and monthly frequencies for a period ranging from January 1997 until November 2014. The empirical results show that high stock volatility explains the flight-to-safety phenomenon during financial stress epochs. Moreover, the findings indicate that the short rate drives the time-variation at daily frequency and that the yield spread and stock liquidity explain the monthly dynamics in stock and government bond return correlations. This research is motivated by its crucial implications on risk management and portfolio optimization. ; Master [120] en Ingénieur de gestion, Université catholique de Louvain, 2015
BASE
Meta-analysis techniques allow researchers to aggregate effect sizes, like standardized regression estimates, of different studies. Recently, continuous-time meta-analysis (CTmeta) has been developed such that the time-interval dependent lagged-parameter estimates can be properly meta-analyzed. This leads to overall standardized lagged-parameter estimates and their multivariate confidence interval. Often, researchers are not only interested in these overall estimates but also in a specific ordering of them: Many researchers have an a priori expectation regarding the ordering of the predictive strength of the cross-lagged relationships; referred to as causal dominance. For example, a researcher might expect, based on literature or expertise, that the lagged relationship between burnout and work engagement is weaker than the reciprocal lagged relationship. Such a hypothesis can be evaluated with an AIC-type theory-based model selection criterion: GORICA. This paper introduces and illustrates how the GORICA can be applied to CTmeta-analyzed standardized lagged-parameter estimates and demonstrate its performance.
BASE