Generalized Forecast Averaging in Autoregressions with a Near Unit Root

2020 ◽  
Author(s):  
Mohitosh Kejriwal ◽  
Xuewen Yu
Keyword(s):  
Time Series ◽  
Least Squares ◽  
Mean Squared Error ◽  
Model Averaging ◽  
Generalized Least Squares ◽  
Ordinary Least Squares ◽  
Finite Sample ◽  
Deterministic Components ◽  
Asymptotic Mean Squared Error ◽  
Near Unit Root

Summary This paper develops a new approach to forecasting a highly persistent time series that employs feasible generalized least squares (FGLS) estimation of the deterministic components in conjunction with Mallows model averaging. Within a local-to-unity asymptotic framework, we derive analytical expressions for the asymptotic mean squared error and one-step-ahead mean squared forecast risk of the proposed estimator and show that the optimal FGLS weights are different from their ordinary least squares (OLS) counterparts. We also provide theoretical justification for a generalized Mallows averaging estimator that incorporates lag order uncertainty in the construction of the forecast. Monte Carlo simulations demonstrate that the proposed procedure yields a considerably lower finite-sample forecast risk relative to OLS averaging. An application to U.S. macroeconomic time series illustrates the efficacy of the advocated method in practice and finds that both persistence and lag order uncertainty have important implications for the accuracy of forecasts.

Nuisance vs. Substance: Specifying and Estimating Time-Series-Cross-Section Models

1996 ◽  
Vol 6 ◽  
pp. 1-36 ◽  
Author(s):  
Nathaniel Beck ◽  
Jonathan N. Katz
Keyword(s):  
Time Series ◽  
Least Squares ◽  
Cross Section ◽  
Generalized Least Squares ◽  
Ordinary Least Squares ◽  
Correlated Errors ◽  
Section Data ◽  
Variable Approach ◽  
Parameter Heterogeneity ◽  
Lagged Dependent Variable

In a previous article we showed that ordinary least squares with panel corrected standard errors is superior to the Parks generalized least squares approach to the estimation of time-series-cross-section models. In this article we compare our proposed method with another leading technique, Kmenta's “cross-sectionally heteroskedastic and timewise autocorrelated” model. This estimator uses generalized least squares to correct for both panel heteroskedasticity and temporally correlated errors. We argue that it is best to model dynamics via a lagged dependent variable rather than via serially correlated errors. The lagged dependent variable approach makes it easier for researchers to examine dynamics and allows for natural generalizations in a manner that the serially correlated errors approach does not. We also show that the generalized least squares correction for panel heteroskedasticity is, in general, no improvement over ordinary least squares and is, in the presence of parameter heterogeneity, inferior to it. In the conclusion we present a unified method for analyzing time-series-cross-section data.

scholarly journals A Combination Method for Averaging OLS and GLS Estimators

Econometrics ◽  
2019 ◽  
Vol 7 (3) ◽  
pp. 38
Author(s):  
Qingfeng Liu ◽  
Andrey L. Vasnev
Keyword(s):  
Least Squares ◽  
Error Term ◽  
Model Averaging ◽  
Generalized Least Squares ◽  
Ordinary Least Squares ◽  
Information Criteria ◽  
Combination Method ◽  
Estimation Accuracy ◽  
Regularity Conditions ◽  
Optimal Weights

To avoid the risk of misspecification between homoscedastic and heteroscedastic models, we propose a combination method based on ordinary least-squares (OLS) and generalized least-squares (GLS) model-averaging estimators. To select optimal weights for the combination, we suggest two information criteria and propose feasible versions that work even when the variance-covariance matrix is unknown. The optimality of the method is proven under some regularity conditions. The results of a Monte Carlo simulation demonstrate that the method is adaptive in the sense that it achieves almost the same estimation accuracy as if the homoscedasticity or heteroscedasticity of the error term were known.

Rates of Expansions for Functional Estimators

2021 ◽  
Author(s):  
Yulia Kotlyarova ◽  
Marcia M. A. Schafgans ◽  
Victoria Zinde-Walsh
Keyword(s):  
Convergence Rates ◽  
Mean Squared Error ◽  
Model Averaging ◽  
Finite Sample ◽  
Squared Error ◽  
Semiparametric Estimators ◽  
Asymptotic Mean Squared Error ◽  
The Impact ◽  
Nonconvex Model

AbstractIn this paper, we summarize results on convergence rates of various kernel based non- and semiparametric estimators, focusing on the impact of insufficient distributional smoothness, possibly unknown smoothness and even non-existence of density. In the presence of a possible lack of smoothness and the uncertainty about smoothness, methods of safeguarding against this uncertainty are surveyed with emphasis on nonconvex model averaging. This approach can be implemented via a combined estimator that selects weights based on minimizing the asymptotic mean squared error. In order to evaluate the finite sample performance of these and similar estimators we argue that it is important to account for possible lack of smoothness.

IMPROVED AND EXTENDED END-OF-SAMPLE INSTABILITY TESTS USING A FEASIBLE QUASI-GENERALIZED LEAST SQUARES PROCEDURE

2009 ◽  
Vol 26 (4) ◽  
pp. 994-1031 ◽  
Author(s):  
Dukpa Kim
Keyword(s):  
Least Squares ◽  
Serial Correlation ◽  
Linear Time ◽  
Generalized Least Squares ◽  
Ordinary Least Squares ◽  
Consistent Estimate ◽  
Linear Regression Models ◽  
Finite Sample ◽  
Error Process ◽  
Economic Statistics

This paper extends the Andrews (2002, Econometrica 71, 1661–1694) and Andrews and Kim (2006, Journal of Business & Economic Statistics 24, 379–394) ordinary least squares–based end-of-sample instability tests for linear regression models. The author proposes to quasi-difference the data first using a consistent estimate of the sum of the autoregressive coefficients of the error process and then test for the end-of-sample instability. For the cointegration model, the feasible quasi-generalized least squares (FQGLS) version of the Andrews and Kim (2006) P test is considered and is shown, by simulations, to be more robust to serial correlation in the error process and to have power no less than Andrews and Kim’s original test. For the linear time trend model, the FQGLS version of the Andrews (2002) S test is considered with the error process allowed to be nonstationary up to one unit root, and the new test is shown to be robust to potentially nonstationary serial correlation. A simulation study also shows that the finite-sample properties of the proposed test can be further improved when the Andrews (1993, Econometrica 61,139–165) or Andrews and Chen (1994, Journal of Business & Economic Statistics 12, 187–204) median unbiased estimate of the sum of the autoregressive coefficients is used.

scholarly journals New Estimator for AR (1) Model with Missing Observations

2021 ◽  
Vol 23 (09) ◽  
pp. 147-159
Author(s):  
Mohamed Khalifa Ahmed Issa ◽  
Keyword(s):  
Time Series ◽  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Least Squares ◽  
Unit Root ◽  
Constant Term ◽  
Ordinary Least Squares ◽  
Missing Observations ◽  
New Form ◽  
Near Unit Root

In this paper, new form of the parameters of AR(1) with constant term with missing observations has been derived by using Ordinary Least Squares (OLS) method, Also, the properties of OLS estimator are discussed, moreover, an extension of Youssef [18]has been suggested for AR(1) with constant with missing observations. A comparative study between (OLS), Yule-Walker (YW) and modification of the ordinary least squares (MOLS) is considered in the case of stationary and near unit root time series, using Monte Carlo simulation.

Minimax robust designs for wavelet estimation of nonparametric regression models with autocorrelated errors

2017 ◽  
Vol 15 (03) ◽  
pp. 1750025
Author(s):  
Selvakkadunko Selvaratnam ◽  
Alwell Julius Oyet
Keyword(s):  
Least Squares ◽  
Nonparametric Regression ◽  
Regression Models ◽  
Simulated Annealing Algorithm ◽  
Mean Squared Error ◽  
Weighted Least Squares ◽  
Generalized Least Squares ◽  
Ordinary Least Squares ◽  
Wavelet Bases ◽  
Autocorrelated Errors

We discuss the construction of designs for estimation of nonparametric regression models with autocorrelated errors when the mean response is to be approximated by a finite order linear combination of dilated and translated versions of the Daubechies wavelet bases with four vanishing moments. We assume that the parameters of the resulting model will be estimated by weighted least squares (WLS) or by generalized least squares (GLS). The bias induced by the unused components of the wavelet bases, in the linear approximation, then inflates the natural variation of the WLS and GLS estimates. We therefore construct our designs by minimizing the maximum value of the average mean squared error (AMSE). Such designs are said to be robust in the minimax sense. Our illustrative examples are constructed by using the simulated annealing algorithm to select an optimal [Formula: see text]-point design, which are integers, from a grid of possible values of the explanatory or design variable [Formula: see text]. We found that the integer-valued designs we constructed based on GLS estimation, have smaller minimum loss when compared to the designs for WLS or ordinary least squares (OLS) estimation, except when the correlation parameter [Formula: see text] approaches 1.

REDUNDANCY OF MOMENT CONDITIONS AND THE EFFICIENCY OF OLS IN SUR MODELS

2008 ◽  
Vol 24 (5) ◽  
pp. 1456-1460 ◽  
Author(s):  
Hailong Qian
Keyword(s):  
Least Squares ◽  
Asymptotic Efficiency ◽  
Generalized Method Of Moments ◽  
Sufficient Conditions ◽  
Generalized Least Squares ◽  
Ordinary Least Squares ◽  
American Statistical Association ◽  
Orthogonality Condition ◽  
Moment Conditions ◽  
Necessary And Sufficient

In this note, based on the generalized method of moments (GMM) interpretation of the usual ordinary least squares (OLS) and feasible generalized least squares (FGLS) estimators of seemingly unrelated regressions (SUR) models, we show that the OLS estimator is asymptotically as efficient as the FGLS estimator if and only if the cross-equation orthogonality condition is redundant given the within-equation orthogonality condition. Using the condition for redundancy of moment conditions of Breusch, Qian, Schmidt, and Wyhowski (1999, Journal of Econometrics 99, 89–111), we then derive the necessary and sufficient condition for the equal asymptotic efficiency of the OLS and FGLS estimators of SUR models. We also provide several useful sufficient conditions for the equal asymptotic efficiency of OLS and FGLS estimators that can be interpreted as various mixings of the two famous sufficient conditions of Zellner (1962, Journal of the American Statistical Association 57, 348–368).

scholarly journals The effect of creative corporate culture and intangibility on the performance of foreign firms traded on the NYSE

2018 ◽  
Vol 15 (4) ◽  
pp. 356-372 ◽  
Author(s):  
Marcia Martins Mendes De Luca ◽  
Paulo Henrique Nobre Parente ◽  
Emanoel Mamede Sousa Silva ◽  
Ravena Rodrigues Sousa
Keyword(s):  
New York ◽  
Least Squares ◽  
Corporate Culture ◽  
Corporate Performance ◽  
Stock Exchange ◽  
Generalized Least Squares ◽  
Ordinary Least Squares ◽  
Foreign Firms ◽  
Content Type ◽  
Negative Effect

Purpose Following the tenets of resource-based view, the present study aims to investigate the effect of creative corporate culture according to the competing values framework model at the level of corporate intangibility and its respective repercussions on performance. Design/methodology/approach The sample included 117 non-USA foreign firms traded on the New York Stock Exchange (NYSE), which issued annual financial reports between 2009 and 2014 using the 20-F form. To meet the study objectives, in addition to the descriptive and comparative analyses, the authors performed regression analyses with panel data, estimating generalized least-squares, two-stage least-squares and ordinary least-squares. Findings Creative culture had a negative effect on the level of intangibility and corporate performance, while the level of intangibility did not appear to influence corporate performance. When combined, creative culture and intangibility had a potentially negative effect on corporate results. In conclusion, creative corporate culture had a negative effect on performance, even in firms with higher levels of intangibility, characterized by elements like experimentation and innovation. Originality/value Although the study hypotheses were eventually rejected, the analyses are relevant to both the academic setting and the market because of the organizational and institutional aspects evaluated, especially in relation to intangibility and creative culture and in view of the unique cross-cultural approach adopted. Within the corporate setting, the study provides a spectrum of stakeholders with tools to identify the profile of foreign firms traded on the NYSE.

Estimating an Autoregressive Current Effects Model of Sales Response when Observations are Aggregated over Time: Least Squares versus Maximum Likelihood

1988 ◽  
Vol 25 (3) ◽  
pp. 301-307
Author(s):  
Wilfried R. Vanhonacker
Keyword(s):  
Maximum Likelihood ◽  
Least Squares ◽  
Serial Correlation ◽  
Mean Squared Error ◽  
Generalized Least Squares ◽  
Parameter Estimates ◽  
Squared Error ◽  
Positive Serial Correlation ◽  
Serial Correlation Coefficient ◽  
Over Time

Estimating autoregressive current effects models is not straightforward when observations are aggregated over time. The author evaluates a familiar iterative generalized least squares (IGLS) approach and contrasts it to a maximum likelihood (ML) approach. Analytic and numerical results suggest that (1) IGLS and ML provide good estimates for the response parameters in instances of positive serial correlation, (2) ML provides superior (in mean squared error) estimates for the serial correlation coefficient, and (3) IGLS might have difficulty in deriving parameter estimates in instances of negative serial correlation.

An Investigation of the Time Series Behaviour of International Tourist Arrivals

1997 ◽  
Vol 3 (2) ◽  
pp. 185-199 ◽  
Author(s):  
Kevin KF. Wong
Keyword(s):  
Time Series ◽  
Least Squares ◽  
Preliminary Investigation ◽  
Model Misspecification ◽  
Theoretical Implication ◽  
Ordinary Least Squares ◽  
Initial State ◽  
International Tourist ◽  
Stationary Type ◽  
Parameter Estimators

Most tourism econometric models are based on conventional least squares estimation, which assumes stationarity in their data generating mechanism. However, they fail to recognize the implications of the integrated properties of the historical time series of tourism data. Such time series properties may have important consequences with regard to the theoretical implication and interpretation of these tourism models. In this paper, historical data on international tourist arrivals from six major regions and seventeen individual countries are analysed to determine whether the series is better characterized by a stationary or non-stationary type process. Based on unit root tests, the results in most cases indicate that international tourist arrivals exhibit a non-stationary stochastic process that has the tendency to fluctuate away from a given initial state as time passes. These findings imply that studies which conveniently draw standard inferences from ordinary least squares estimated tourism models based on the levels of international tourist arrivals can be very misleading since non-stationarity in the data will produce inconsistent parameter estimators and unreliable test statistics. Furthermore, model misspecification that arises from unrelated integrated series can seriously bias conventional significance tests towards the acceptance of an apparently significant relationship. In this preliminary investigation, we conclude that econometric tourism models that focus on the levels of international tourist arrivals may not be reliable since the series is non-stationary and is integrated of order one, I(1).

Sign in / Sign up

Export Citation Format

Share Document