umu.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Publications (6 of 6) Show all publications
Forchini, G. & Jiang, B. (2019). The unconditional distributions of the OLS, TSLS and LIML estimators in a simple structural equations model. Econometric Reviews, 38(2), 208-247
Open this publication in new window or tab >>The unconditional distributions of the OLS, TSLS and LIML estimators in a simple structural equations model
2019 (English)In: Econometric Reviews, ISSN 0747-4938, E-ISSN 1532-4168, Vol. 38, no 2, p. 208-247Article in journal (Refereed) Published
Abstract [en]

The exact distributions of the standard estimators of the structural coefficients in a linear structural equations model conditional on the exogenous variables have been shown to have some unexpected and quirky features. Since the argument for conditioning on exogenous (ancillary) variables has been weakened over the past 20 years by the discovery of an “ancillarity paradox,” it is natural to wonder whether such finite sample properties are in fact due to conditioning on the exogenous variables. This article studies the exact distributions of the ordinary least squares (OLS), two-stage least squares (TSLS), and limited information maximum likelihood (LIML) estimators of the structural coefficients in a linear structural equation without conditioning on the exogenous variables.

Place, publisher, year, edition, pages
Philadelphia: Taylor & Francis, 2019
Keywords
Exact distribution, limited information maximum likelihood, linear structural equations, ordinary least squares, two-stage least squares
National Category
Economics and Business
Research subject
Econometrics
Identifiers
urn:nbn:se:umu:diva-131480 (URN)10.1080/07474938.2016.1261072 (DOI)000465616100005 ()
Available from: 2017-02-15 Created: 2017-02-15 Last updated: 2019-05-27Bibliographically approved
Forchini, G., Jiang, B. & Peng, B. (2018). TSLS and LIML Estimators in Panels with Unobserved Shocks. Econometrics, 6(2), Article ID 19.
Open this publication in new window or tab >>TSLS and LIML Estimators in Panels with Unobserved Shocks
2018 (English)In: Econometrics, ISSN 2225-1146, Vol. 6, no 2, article id 19Article in journal (Refereed) Published
Abstract [en]

The properties of the two stage least squares (TSLS) and limited information maximum likelihood (LIML) estimators in panel data models where the observables are affected by common shocks, modelled through unobservable factors, are studied for the case where the time series dimension is fixed. We show that the key assumption in determining the consistency of the panel TSLS and LIML estimators, as the cross section dimension tends to infinity, is the lack of correlation between the factor loadings in the errors and in the exogenous variables-including the instruments-conditional on the common shocks. If this condition fails, both estimators have degenerate distributions. When the panel TSLS and LIML estimators are consistent, they have covariance-matrix mixed-normal distributions asymptotically. Tests on the coefficients can be constructed in the usual way and have standard distributions under the null hypothesis.

Keywords
two-stage least squares, limited information maximum likelihood, common shocks
National Category
Economics
Identifiers
urn:nbn:se:umu:diva-150868 (URN)10.3390/econometrics6020019 (DOI)000436274600005 ()
Available from: 2018-09-03 Created: 2018-09-03 Last updated: 2018-09-05Bibliographically approved
Forchini, G. & Peng, B. (2017). A modified first difference estimator for panel data models with a multifactor error structure when the time dimension is small. Communications in Statistics, forthcoming
Open this publication in new window or tab >>A modified first difference estimator for panel data models with a multifactor error structure when the time dimension is small
2017 (English)In: Communications in Statistics, Vol. forthcomingArticle in journal (Refereed) In press
National Category
Economics and Business
Research subject
Econometrics
Identifiers
urn:nbn:se:umu:diva-131478 (URN)
Available from: 2017-02-15 Created: 2017-02-15 Last updated: 2019-01-15
Forchini, G. & Peng, B. (2017). Modified first-difference estimator in a panel data model with unobservable factors both in the errors and the regressors when the time dimension is small. Communications in Statistics - Theory and Methods, 46(24), 12226-12239
Open this publication in new window or tab >>Modified first-difference estimator in a panel data model with unobservable factors both in the errors and the regressors when the time dimension is small
2017 (English)In: Communications in Statistics - Theory and Methods, ISSN 0361-0926, E-ISSN 1532-415X, Vol. 46, no 24, p. 12226-12239Article in journal (Refereed) Published
Abstract [en]

Panel data models with factor structures in both the errors and the regressors have received considerable attention recently. In these models, the errors and the regressors are correlated and the standard estimators are inconsistent. This paper shows that, for such models, a modified first-difference estimator (in which the time and the cross-sectional dimensions are interchanged) is consistent as the cross-sectional dimension grows but the time dimension is small. Although the estimator has a non standard asymptotic distribution, t and F tests have standard asymptotic distribution under the null hypothesis.

Place, publisher, year, edition, pages
Taylor & Francis, 2017
Keywords
common shocks, factors, panel data
National Category
Probability Theory and Statistics
Identifiers
urn:nbn:se:umu:diva-142678 (URN)10.1080/03610926.2017.1295156 (DOI)000415803800018 ()
Available from: 2017-12-11 Created: 2017-12-11 Last updated: 2018-06-09Bibliographically approved
Forchini, G. & Peng, B. (2016). A conditional approach to panel data models with common shocks. Econometrics, 4(1), Article ID 4.
Open this publication in new window or tab >>A conditional approach to panel data models with common shocks
2016 (English)In: Econometrics, ISSN 2225-1146, Vol. 4, no 1, article id 4Article in journal (Refereed) Published
Abstract [en]

This paper studies the effects of common shocks on the OLS estimators of the slopes' parameters in linear panel data models. The shocks are assumed to affect both the errors and some of the explanatory variables. In contrast to existing approaches, which rely on using results on martingale difference sequences, our method relies on conditional strong laws of large numbers and conditional central limit theorems for conditionally-heterogeneous random variables.

Keywords
factor structure, common shocks, conditional independence, conditional central limit theorem
National Category
Economics
Identifiers
urn:nbn:se:umu:diva-131233 (URN)10.3390/econometrics4010004 (DOI)
Available from: 2017-02-09 Created: 2017-02-09 Last updated: 2018-06-09Bibliographically approved
Forchini, G. & Ranasinghe, K. (2016). Modelling Multivariate Durations. International Journal of Statistics & Economic, 17(1), 82-93
Open this publication in new window or tab >>Modelling Multivariate Durations
2016 (English)In: International Journal of Statistics & Economic, ISSN 0975-556X, Vol. 17, no 1, p. 82-93Article in journal (Refereed) Published
Abstract [en]

We propose a simple procedure to model multivariate durations. This is done by specifying two conditional models: anautoregressive conditional durationmodel for on a pooled series of durationsand a logit model for the type marks. We estimate the model by maximising a pseudo-likelihood which is equivalent to estimating the autoregressive conditional duration model and the logit model separately. We illustrate this methodology by modelling the joint dynamics of the trade shares of Tabcorp Holdings Limited and Tatts group Limited in the Australian financial market between January 15 and January 31 2009.

Keywords
High-frequency data, autoregressive conditional durations, multivariate durations, multinomial logit, conditional distributions
National Category
Economics
Identifiers
urn:nbn:se:umu:diva-131232 (URN)
Available from: 2017-02-09 Created: 2017-02-09 Last updated: 2018-06-09Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0002-1377-9469

Search in DiVA

Show all publications