Modeling the Autocorrelated Errors in Time Series Regression: A Generalized Least Squares Approach

Akpan, Emmanuel Alphonsus; Moffat, Imoh Udo

doi:10.9734/jamcs/2018/39949

Cited by 8 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…12 The model was then reestimated using a generalized least squares approach, which allows possible unequal variances and include an autoregressive moving average process to model the autocorrelation. 13 Results of the final model are presented in tables with 95% confidence intervals (95% CIs).…”

Section: Discussionmentioning

confidence: 99%

Effect of regulatory interventions on agomelatine use in Spain: A multiple intervention time‐series analysis in a nationwide electronic healthcare record database

González-Bermejo

Castillo-Cano

Rodríguez-Pascual

et al. 2021

Pharmacoepidemiology and Drug

View full text Add to dashboard Cite

Background: Liver injury is an important identified risk for agomelatine and several measures were put in place to prevent and minimize such risk. The study aims to assess the impact of four interventions on the incidence of agomelatine use, particularly among patients aged ≥75 in Spain between 2011 and 2018.Methods: Quasi-experimental interrupted time-series analysis to examine data from a nationwide electronic healthcare record database (BIFAP). Quarterly cumulative incidence of agomelatine use per 100 000 patients was calculated and the impact of four regulatory interventions was quantified. Results:The incidence of agomelatine use decreased by 85% and 87% from first quarter 2011 to last quarter 2018 in patients below and above 75 years old, respectively. Regulatory actions taken were not associated with an immediate and significant falling level of use or slope. The incidence was less than expected 6 months after the first and third intervention for patients below and above 75 years old, and more than expected after the second and fourth intervention for both populations, though these analyses were underpowered to observe significant results. The downward trend became less pronounced, reaching a residual level of use, which remained stable in the last segment of the study period. Conclusion: New users of agomelatine decreased throughout the study period, starting before interventions took place. The effect of specific interventions might be masked by the progressive decrease tendency, constant over the study period. The effects of external factors that might overlap, unintended consequences, and issues concerning statistical modeling in situations where rates are already falling, should be considered when interpreting the results.

show abstract

Section: Discussionmentioning

confidence: 99%

Effect of regulatory interventions on agomelatine use in Spain: A multiple intervention time‐series analysis in a nationwide electronic healthcare record database

González-Bermejo

Castillo-Cano

Rodríguez-Pascual

et al. 2021

Pharmacoepidemiology and Drug

View full text Add to dashboard Cite

show abstract

“…As indicated by the studies of Akpan and Moffat (2018) and Gopinath, Krishnamurthy, and Sathian (2018), the FEM technique can be used to resolve issues of autocorrelation and heteroscedasticity. Hence, the FEM analysis technique is undertaken to address the heteroscedasticity observed in the preliminary testing.…”

Section: Fixed Effect and Robust Standard Error Regression Analysismentioning

confidence: 99%

Educational Mismatch and Income Inequality in Developing Countries

Tan

Wang

2021

IJSSP

View full text Add to dashboard Cite

This study aims to evaluate the influence of educational mismatch on income inequality in developing countries. The research is conducted using secondary data from ten developing countries over a twenty-year period (2001–2020). Statistical analysis is performed on the dataset using Stata, where the main variable representing income inequality is the GINI coefficient, and educational mismatch is assessed using educational attainment, educational attainment of females, and educational expenditure. The fixed effect and robust standard error techniques are applied to address the association among the variables. The study's findings reveal that educational attainment and educational expenditure have a positive influence on the GINI coefficient, indicating that the two factors contribute to an educational mismatch in the developing country. On the other hand, the findings also reveal that the educational attainment of females has a negative effect on the GINI coefficient, indicating that a high level of educational attainment among females can help to improve the income equality of developing countries. The implications of the study are particularly relevant to the policymakers and governments of developing countries. They should take steps to increase the level of education among females to contribute to increased income equality levels in their country.

show abstract

“…They further pointed out that autocorrelation leads to a situation where the predicted values are too high for some periods and too low for others and thus, a series of negative residuals alternate with a series of positive residuals. Akpan and Moffat (2018) opined that if the assumption of no correlation in the error term is violated, then, the underlying model would be rendered invalid with the standard errors of the parameters becoming biased. Moreover, if the errors are correlated, the least squares estimators are inefficient and the estimated variances were not appropriate.…”

Section: Introductionmentioning

confidence: 99%

Bayesian Generalized Least Squares with Autocorrelated Error

Oloyede

2022

JOSMA

View full text Add to dashboard Cite

Autocorrelation plays significant role in both time series and cross sectional data. More often than none it rendered the inference of parameter estimates invalid and those other statistics that use the parameters. This study investigates the asymptotic behaviour of generalised least squares with Autocorrelated errors cum Marcov Chain Monte Carlo simulation. Bias and Mean Squares Error criteria were used to evaluate finite properties of the estimator. The following sample sizes: 25, 50,100, and 250 were constructed and used. Thus 11,000 simulations with varying level of Autocorrelated error were carried out. This is subjected to the level of convergence. Bias and Minimum Mean Squares Error criteria revealed improving performance asymptotically regardless of the level of Autocorrelated error. The study observed that asymptotic consistency and efficiency were obtained at large sample which obey the law of large number and point to the fact that variance of error terms tend towards zero and distribution tends to normal when law of large number is applied. The study therefore recommended that large samples should be obtain to make the inferences stable

show abstract

Modeling the Autocorrelated Errors in Time Series Regression: A Generalized Least Squares Approach

Cited by 8 publications

References 0 publications

Effect of regulatory interventions on agomelatine use in Spain: A multiple intervention time‐series analysis in a nationwide electronic healthcare record database

Effect of regulatory interventions on agomelatine use in Spain: A multiple intervention time‐series analysis in a nationwide electronic healthcare record database

Educational Mismatch and Income Inequality in Developing Countries

Bayesian Generalized Least Squares with Autocorrelated Error

Contact Info

Product

Resources

About