Results 1 - 10
of
243
Estimating standard errors in finance panel data sets: comparing approaches.
- Review of Financial Studies
, 2009
"... Abstract In both corporate finance and asset pricing empirical work, researchers are often confronted with panel data. In these data sets, the residuals may be correlated across firms and across time, and OLS standard errors can be biased. Historically, the two literatures have used different solut ..."
Abstract
-
Cited by 890 (7 self)
- Add to MetaCart
(Show Context)
Abstract In both corporate finance and asset pricing empirical work, researchers are often confronted with panel data. In these data sets, the residuals may be correlated across firms and across time, and OLS standard errors can be biased. Historically, the two literatures have used different solutions to this problem. Corporate finance has relied on clustered standard errors, while asset pricing has used the Fama-MacBeth procedure to estimate standard errors. This paper examines the different methods used in the literature and explains when the different methods yield the same (and correct) standard errors and when they diverge. The intent is to provide intuition as to why the different approaches sometimes give different answers and give researchers guidance for their use.
How much should we trust differences-in-differences estimates?
, 2003
"... Most papers that employ Differences-in-Differences estimation (DD) use many years of data and focus on serially correlated outcomes but ignore that the resulting standard errors are inconsistent. To illustrate the severity of this issue, we randomly generate placebo laws in state-level data on femal ..."
Abstract
-
Cited by 828 (1 self)
- Add to MetaCart
(Show Context)
Most papers that employ Differences-in-Differences estimation (DD) use many years of data and focus on serially correlated outcomes but ignore that the resulting standard errors are inconsistent. To illustrate the severity of this issue, we randomly generate placebo laws in state-level data on female wages from the Current Population Survey. For each law, we use OLS to compute the DD estimate of its “effect” as well as the standard error of this estimate. These conventional DD standard errors severely understate the standard deviation of the estimators: we find an “effect ” significant at the 5 percent level for up to 45 percent of the placebo interventions. We use Monte Carlo simulations to investigate how well existing methods help solve this problem. Econometric corrections that place a specific parametric form on the time-series process do not perform well. Bootstrap (taking into account the auto-correlation of the data) works well when the number of states is large enough. Two corrections based on asymptotic approximation of the variance-covariance matrix work well for moderate numbers of states and one correction that collapses the time series information into a “pre” and “post” period and explicitly takes into account the effective sample size works well even for small numbers of states.
Robust Inference with Multi-way Clustering
, 2006
"... In this paper we propose a new variance estimator for OLS as well as for nonlinear estimators such as logit, probit and GMM. This variance estimator enables cluster-robust inference when there is two-way or multi-way clustering that is nonnested. The variance estimator extends the standard cluster-r ..."
Abstract
-
Cited by 363 (4 self)
- Add to MetaCart
In this paper we propose a new variance estimator for OLS as well as for nonlinear estimators such as logit, probit and GMM. This variance estimator enables cluster-robust inference when there is two-way or multi-way clustering that is nonnested. The variance estimator extends the standard cluster-robust variance estimator or sandwich estimator for one-way clustering (e.g. Liang and Zeger (1986), Arellano (1987)) and relies on similar relatively weak distributional assumptions. Our method is easily implemented in statistical packages, such as Stata and SAS, that already offer cluster-robust standard errors when there is one-way clustering. The method is demonstrated by a Monte Carlo analysis for a two-way random effects model; a Monte Carlo analysis of a placebo law that extends the state-year effects example of Bertrand et al. (2004) to two dimensions; and by application to two studies in the empirical public/labor literature where two-way clustering is present.
Five Facts About Prices: A Reevaluation of Menu Cost Models,” Quarterly
- Journal of Economics
, 2008
"... Abstract We establish five facts about prices in the U.S. economy: 1) The median frequency of nonsale price change is 9-12% per month, roughly half of what it is including sales. This implies an uncensored median duration of regular prices of 8-11 months. Product turnover plays an important role in ..."
Abstract
-
Cited by 326 (9 self)
- Add to MetaCart
Abstract We establish five facts about prices in the U.S. economy: 1) The median frequency of nonsale price change is 9-12% per month, roughly half of what it is including sales. This implies an uncensored median duration of regular prices of 8-11 months. Product turnover plays an important role in truncating price spells in durable goods. The median frequency of price change for finished goods producer prices is roughly 11% per month. 2) One-third of regular price changes are price decreases. 3) The frequency of price increases covaries strongly with inflation while the frequency of price decreases and the size of price increases and price decreases do not. 4) The frequency of price change is highly seasonal: It is highest in the 1st quarter and lowest in the 4th quarter. 5) The hazard function of price changes for individual consumer and producer goods is downward sloping for the first few months and then flat (except for a large spike at 12 months in consumer services and all producer prices). These facts are based on CPI microdata and a new comprehensive data set of microdata on producer prices that we construct from raw production files underlying the PPI. We show that the 1st, 2nd and 3rd facts are consistent with a benchmark menu-cost model, while the 4th and 5th facts are not.
Bootstrap-Based Improvements for Inference with Clustered Errors
, 2006
"... Microeconometrics researchers have increasingly realized the essential need to account for any within-group dependence in estimating standard errors of regression parameter estimates. The typical preferred solution is to calculate cluster-robust or sandwich standard errors that permit quite general ..."
Abstract
-
Cited by 303 (12 self)
- Add to MetaCart
Microeconometrics researchers have increasingly realized the essential need to account for any within-group dependence in estimating standard errors of regression parameter estimates. The typical preferred solution is to calculate cluster-robust or sandwich standard errors that permit quite general heteroskedasticity and within-cluster error correlation, but presume that the number of clusters is large. In applications with few (5-30) clusters, standard asymptotic tests can overreject considerably. We investigate more accurate inference using cluster bootstrap-t procedures that provide asymptotic refinement. These procedures are evaluated using Monte Carlos, including the much-cited differences-in-differences example of Bertrand, Mullainathan and Duflo (2004). In situations where standard methods lead to rejection rates in excess of ten percent (or more) for tests of nominal size 0.05, our methods can reduce this to five percent. In principle a pairs cluster bootstrap should work well, but in practice a wild cluster bootstrap performs better.
Instrumental variables and GMM: Estimation and testing
- Stata Journal
, 2003
"... Abstract. We discuss instrumental variables (IV) estimation in the broader context of the generalized method of moments (GMM), and describe an extended IV estimation routine that provides GMM estimates as well as additional diagnostic tests. Stand–alone test procedures for heteroskedasticity, overid ..."
Abstract
-
Cited by 192 (7 self)
- Add to MetaCart
Abstract. We discuss instrumental variables (IV) estimation in the broader context of the generalized method of moments (GMM), and describe an extended IV estimation routine that provides GMM estimates as well as additional diagnostic tests. Stand–alone test procedures for heteroskedasticity, overidentification, and endogeneity in the IV context are also described.
The enrollment effects of merit-based financial aid: Evidence from Georgia's HOPE scholarship
- Journal of Labor Economics
, 2006
"... and two anonymous referees for helpful comments and suggestions. Cornwell and Mustard gratefully acknowledge ..."
Abstract
-
Cited by 89 (6 self)
- Add to MetaCart
and two anonymous referees for helpful comments and suggestions. Cornwell and Mustard gratefully acknowledge
The Politics of Foreign Direct Investment into Developing Countries
- Increasing FDI through International Trade Agreements?” American Journal of Political Science 52(4):741–762
, 2008
"... The flow of foreign direct investment into developing countries varies greatly across countries and over time. The political factors that affect these flows are not well understood. Focusing on the relationship between trade and investment, we argue that international trade agreements—GATT/WTO and p ..."
Abstract
-
Cited by 77 (2 self)
- Add to MetaCart
The flow of foreign direct investment into developing countries varies greatly across countries and over time. The political factors that affect these flows are not well understood. Focusing on the relationship between trade and investment, we argue that international trade agreements—GATT/WTO and preferential trade agreements (PTAs)—provide mechanisms for making commitments to foreign investors about the treatment of their assets, thus reassuring investors and increasing investment. These international commitments are more credible than domestic policy choices, because reneging on them is more costly. Statistical analyses for 122 developing countries from 1970 to 2000 support this argument. Developing countries that belong to the WTO and participate in more PTAs experience greater FDI inflows than otherwise, controlling for many factors including domestic policy preferences and taking into account possible endogeneity. Joining international trade agreements allows developing countries to attract more FDI and thus increase economic growth. Foreign direct investment (FDI) by multinationalcorporations (MNCs) has grown rapidly in recentdecades,1 and developing countries have attracted an increasing share of it: $334 billion in 2005, or more than 36 % of all inward FDI flows (UNCTAD 2006, xvii). Its importance for developing countries ’ economies also has increased, from an average of barely 1 % of GDP in the 1970s to about 2.5 % of GDP on average by 2000. Yet,
The productivity effects of privatization: Longitudinal estimates from
- and Ukraine,” Journal of Political Economy
, 2006
"... This paper estimates the effect of privatization on multifactor productivity (MFP) using long panel data for nearly the universe of initially state-owned manufacturing firms in four economies. We exploit the key longitudinal feature of our data to measure and control for pre-privatization selection ..."
Abstract
-
Cited by 45 (7 self)
- Add to MetaCart
This paper estimates the effect of privatization on multifactor productivity (MFP) using long panel data for nearly the universe of initially state-owned manufacturing firms in four economies. We exploit the key longitudinal feature of our data to measure and control for pre-privatization selection bias and to estimate long-run impacts. We find that the magnitudes of our estimates are robust to alternative functional forms, but sensitive to how we control for selection. Our preferred random growth models imply that majority privatization raises MFP about 15 % in
