Using Bootstrapped Confidence Intervals for Improved Inferences with Seemingly Unrelated Regression Equations

Rilstone, Paul; Veall, Michael R.

doi:10.1017/s026646660000685x

Cited by 27 publications

(22 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Rilstone and Veall (1996) provide encouraging evidence on the performance of certain bootstrap procedures in the context of seemingly unrelated regressions, and Inoue and Kilian (2002) do so in the context of vector autoregressions. But in neither case, and even less so for simultaneous equations models, should we expect the sort of astonishingly good performance that was observed in Figure 3.…”

Section: Simultaneous Equations Modelsmentioning

confidence: 97%

Estimation and Inference in Econometrics.

Davidson¹,

Davidson²,

MacKinnon³

1995

Economica

1,167

1,293

View full text Add to dashboard Cite

The astonishing increase in computer performance over the past two decades has made it possible for economists to base many statistical inferences on simulated, or bootstrap, distributions rather than on distributions obtained from asymptotic theory. In this paper, I review some of the basic ideas of bootstrap inference. The paper discusses Monte Carlo tests, several types of bootstrap test, and bootstrap confidence intervals. Although bootstrapping often works well, it does not do so in every case.

show abstract

Section: Simultaneous Equations Modelsmentioning

confidence: 97%

Estimation and Inference in Econometrics.

Davidson¹,

Davidson²,

MacKinnon³

1995

Economica

1,167

1,293

View full text Add to dashboard Cite

show abstract

“…However, Wald tests face a number of statistical problems; in particular, (i) the tendency to over-reject the null hypothesis in system OLS and SUR models, and (ii) the non-invariance of nonlinear Wald test statistics to the mathematical reformulation of the null. To address the first issue, we estimate the appropriate p-value of the test statistic using the percentile-F interval of the statistic based on its bootstrap distribution, which has been shown to significantly reduce the over-rejection bias in this setting (Rilstone and Veall 1996). 21 To address the second issue, we assess the robustness of our inferences by constructing linear Wald tests based on the system of parametric versions of the z-conditional demand functions, which we discuss below.…”

Section: Tests Based On Standard (Unconditional) Demand System Estimamentioning

confidence: 99%

Is the Allocation of Resources within the Household Efficient? New Evidence from a Randomized Experiment

Bobonis

2009

Journal of Political Economy

234

147

View full text Add to dashboard Cite

This paper studies whether households make Pareto-efficient intra-household resource allocation decisions. It sets out a new empirical test of the collective model that exploits exogenous variation in two distribution factors -variables affecting the resource sharing rule within households. Such variation is necessary to test the condition that consumption good demand responses are proportional to changes in the distribution of decision-makers' resources. Combining randomized variation in women's income generated by the experimental evaluation of the PROGRESA program in rural Mexico with general household income variation attributable to localized rainfall shocks, we find evidence favoring the Pareto-efficiency condition. More specifically, we find that female-specific income changes have a substantial positive effect on food expenditures and expenditures on children's goods, whereas income changes due to rainfall shocks have a smaller influence on household public goods expenditures. The evidence is consistent with female partners having greater marginal willingness-to-pay sensitivity to own-income changes, and social norms which may oblige women to devote their earnings to meet collective consumption needs.

show abstract

“…Atkinson and Wilson, 1992) have shown that bootstrapped standard errors for the SUR model are also biased downwards and therefore cannot improve the situation. However in a recent paper Rilestone and Veall (1996) present Monte Carlo evidence that the bootstrap performs substantially better when applied to t-ratios (i.e. utilizing the so called percentile-t bootstrap) rather than to standard errors.…”

Section: Kh Ghwhuplqdqwv Ri H[fkdqjh Udwh H[srvxuh Dqg Wkh Lpsdfw Rimentioning

confidence: 99%

Exchange rate exposure, foreign involvement and currency hedging of firms: some Swedish evidence

Nydahl¹

1999

Euro Fin Management

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may $EVWUDFWThis paper investigates the effect of exchange rate fluctuation on a firm's value, the so-called exchange rate exposure, for a sample of Swedish firms. In contrast to previous results, using U.S. data, the values of Swedish firms, as reflected in the stock price, seem quite sensitive to movements in the exchange rate. Studying the cross sectional differences in exposure, the estimated exposure is positively and significantly related to the fractional of total sales made abroad and negatively related to the use of currency derivatives.

show abstract

Using Bootstrapped Confidence Intervals for Improved Inferences with Seemingly Unrelated Regression Equations

Cited by 27 publications

References 13 publications

Estimation and Inference in Econometrics.

Estimation and Inference in Econometrics.

Is the Allocation of Resources within the Household Efficient? New Evidence from a Randomized Experiment

Exchange rate exposure, foreign involvement and currency hedging of firms: some Swedish evidence

Contact Info

Product

Resources

About