Results 1  10
of
328
How much should we trust differencesindifferences estimates? Quarterly Journal of Economics 119:249–75
, 2004
"... Most papers that employ DifferencesinDifferences 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 statelevel data on fema ..."
Abstract

Cited by 775 (1 self)
 Add to MetaCart
Most papers that employ DifferencesinDifferences 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 statelevel 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 timeseries process do not perform well. Bootstrap (taking into account the autocorrelation of the data) works well when the number of states is large enough. Two corrections based on asymptotic approximation of the variancecovariance 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.
The nature of precautionary wealth
 Journal of Monetary Economics
, 1997
"... This paper uses the Panel Study of Income Dynamics to provide some of the first direct evidence that wealth is systematically higher for consumers with predictably greater income uncertainty. However, the apparent pattern of precautionary wealth is not consistent with a standard parameterization of ..."
Abstract

Cited by 273 (17 self)
 Add to MetaCart
This paper uses the Panel Study of Income Dynamics to provide some of the first direct evidence that wealth is systematically higher for consumers with predictably greater income uncertainty. However, the apparent pattern of precautionary wealth is not consistent with a standard parameterization of the life cycle model in which consumers are patient enough to begin saving for retirement early in life: wealth is estimated to be far less sensitive to uncertainty than implied by that model. Instead, our results suggest that over most of their working lifetime, consumers behave in accordance with the "bufferstock" models of saving described in Carroll (1992, 1997) or Deaton (1991), in which consumers hold wealth principally to insulate consumption against nearterm fluctuations in income. JEL Classification: D91, E21
INCOME VARIANCE DYNAMICS AND HETEROGENEITY
"... Recent theoretical work has shown the importance of measuring microeconomic uncertainty for models of both general and partial equilibrium under imperfect insurance. In this paper the assumption of i.i.d. income innovations used in previous empirical studies is removed and the focus of the analysis ..."
Abstract

Cited by 231 (21 self)
 Add to MetaCart
Recent theoretical work has shown the importance of measuring microeconomic uncertainty for models of both general and partial equilibrium under imperfect insurance. In this paper the assumption of i.i.d. income innovations used in previous empirical studies is removed and the focus of the analysis placed on models for the conditional variance of income shocks, which is related to the measure of risk emphasized by the theory. We first discriminate amongst various models of earnings determination that separate income shocks into idiosyncratic transitory and permanent components. We allow for education and timespecific differences in the stochastic process for earnings and for measurement error. The conditional variance of the income shocks is modelled as a parsimonious ARCH process with both observable and unobserved heterogeneity. The empirical analysis is conducted on data drawn from the 19671992 Panel Study of Income Dynamics. We find strong evidence of sizeable ARCH effects as well as evidence of unobserved heterogeneity in the variances.
An Analysis of Sample Attrition in Panel Data. The Michigan Panel Study on Income Dynamics
 Journal of Human Resources
, 1998
"... experienced approximately 50 percent sample loss from cumulative attrition from its initial 1968 membership. We study the effect of this attrition on the unconditional distributions of several socioeconomic variables and on the estimates of several sets of regression coefficients. We provide a stati ..."
Abstract

Cited by 195 (7 self)
 Add to MetaCart
(Show Context)
experienced approximately 50 percent sample loss from cumulative attrition from its initial 1968 membership. We study the effect of this attrition on the unconditional distributions of several socioeconomic variables and on the estimates of several sets of regression coefficients. We provide a statistical framework for conducting tests for attrition bias that draws a sharp distinction between selection on unobservables and on observables and that shows that weighted least squares can generate consistent parameter estimates when selection is based on observables, even when they are endogenous. Our empirical analysis shows that attrition is highly selective and is concentrated among lower socioeconomic status individuals. We also show that attrition is concentrated among those with more unstable earnings, marriage, and migration histories. Nevertheless, we find that these variables explain very little of the attrition in the sample, and that the selection that occurs is moderated by regressiontothemean effects
Inattentive consumers
 Journal of Monetary Economics
, 2006
"... This paper studies the consumption decisions of agents who face costs of acquiring, absorbing and processing information. These consumers rationally choose to only sporadically update their information and recompute their optimal consumption plans. In between updating dates, they remain inattentive ..."
Abstract

Cited by 178 (13 self)
 Add to MetaCart
This paper studies the consumption decisions of agents who face costs of acquiring, absorbing and processing information. These consumers rationally choose to only sporadically update their information and recompute their optimal consumption plans. In between updating dates, they remain inattentive. This behavior implies that news disperses slowly throughout the population, so events have a gradual and delayed effect on aggregate consumption. The model predicts that aggregate consumption adjusts slowly to shocks, and is able to explain the excess sensitivity and excess smoothness puzzles. In addition, individual consumption is sensitive to ordinary and unexpected past news, but it is not sensitive to extraordinary or predictable events. The model further predicts that some people rationally choose to not plan, live handtomouth, and save less, while other people sporadically update their plans. The longer are these plans, the more they save. Evidence using U.S. aggregate and microeconomic data generally supports these predictions.
Earnings dynamics and inequality among Canadian men. NBER Working Paper Series, Working Paper 7370
, 1999
"... Several recent studies have found that earnings inequality in Canada has grown considerably since the late 1970's. Using an extraordinary data base drawn from longitudinal income tax records, we decompose this growth in earnings inequality into its persistent and transitory components. We find ..."
Abstract

Cited by 169 (2 self)
 Add to MetaCart
(Show Context)
Several recent studies have found that earnings inequality in Canada has grown considerably since the late 1970's. Using an extraordinary data base drawn from longitudinal income tax records, we decompose this growth in earnings inequality into its persistent and transitory components. We find that the growth in earnings inequality reflects both an increase in longrun inequality and an increase in earnings instability. The large size of our earnings panel allows us to estimate and test richer models of earnings dynamics than could be supported by the relatively small panel surveys used in U.S. research. The Canadian data strongly reject several restrictions commonly imposed in the U.S. literature, and they also suggest that imposing these evidently false restrictions may lead to distorted inferences about earnings
How often to sample a continuoustime process in the presence of market microstructure noise
 Review of Financial Studies
, 2005
"... In theory, the sum of squares of log returns sampled at high frequency estimates their variance. When market microstructure noise is present but unaccounted for, however, we show that the optimal sampling frequency is finite and derives its closedform expression. But even with optimal sampling, usi ..."
Abstract

Cited by 159 (13 self)
 Add to MetaCart
In theory, the sum of squares of log returns sampled at high frequency estimates their variance. When market microstructure noise is present but unaccounted for, however, we show that the optimal sampling frequency is finite and derives its closedform expression. But even with optimal sampling, using say 5min returns when transactions are recorded every second, a vast amount of data is discarded, in contradiction to basic statistical principles. We demonstrate that modeling the noise and using all the data is a better solution, even if one misspecifies the noise distribution. So the answer is: sample as often as possible. Over the past few years, price data sampled at very high frequency have become increasingly available in the form of the Olsen dataset of currency exchange rates or the TAQ database of NYSE stocks. If such data were not affected by market microstructure noise, the realized volatility of the process (i.e., the average sum of squares of logreturns sampled at high frequency) would estimate the returns ’ variance, as is well known. In fact, sampling as often as possible would theoretically produce in the limit a perfect estimate of that variance. We start by asking whether it remains optimal to sample the price process at very high frequency in the presence of market microstructure noise, consistently with the basic statistical principle that, ceteris paribus, more data are preferred to less. We first show that, if noise is present but unaccounted for, then the optimal sampling frequency is finite, and we We are grateful for comments and suggestions from the editor, Maureen O’Hara, and two anonymous
Learning your Earning: Are Labor Income Shocks Really Very Persistent?
 ECONOMETRICA
, 2004
"... In this paper we examine the risk situation facing individuals in the labor market. The current consensus in the literature is that the labor income process has a large random walk component. We argue two points. First, the estimates of persistence from income data appear to be upward biased due to ..."
Abstract

Cited by 154 (10 self)
 Add to MetaCart
In this paper we examine the risk situation facing individuals in the labor market. The current consensus in the literature is that the labor income process has a large random walk component. We argue two points. First, the estimates of persistence from income data appear to be upward biased due to the omission of heterogeneity in income profiles across the population that would be implied, for example, by a human capital model with heterogeneity. When we allow for differences in profiles, the estimated persistence falls from 0.99 to about 0.8. Moreover, the main evidence against pro…le heterogeneity in the existing literature — that the autocorrelations of income changes are small and typically negative — is also replicated by the pro…le heterogeneity model we estimate, casting doubt on the previous interpretation of this evidence. Second, we embed this process in a lifecycle model to examine how it alters individuals’ consumptionsaving decision. We assume that — as seems plausible — individuals do not know their profiles exactly at the beginning of life, but learn in a Bayesian way with successive income observations. We find that learning is very slow and affects consumption decision throughout the lifecycle. The model generates substantial rise