Monte Carlo evidence on panel data regressions with AR(1) disturbances and an arbitrary variance on the initial observations

Baltagi, Badi H.; Chang, Young Jae; Li, Qi

doi:10.1016/0304-4076(92)90017-l

Cited by 14 publications

(6 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Its relative e ciency appears to improve as the sample size increases, but even when N = 100 and T = 10, its RMSE is usually over one and a half times the true GLS estimator's RMSE. These results mirror the poor showing of the Cochrane-Orcutt estimator in other contexts (see, e.g., Baltagi et al, 1992;Beach and MacKinnon, 1978;Maeshiro, 1976Maeshiro, , 1979Park and Mitchell, 1980;Thornton, 1987). 10 As expected, the two feasible GLS estimators improve upon the Cochrane-Orcutt and OLS estimators in terms of e ciency.…”

Section: Monte Carlo Results For Normal Componentssupporting

confidence: 79%

“…And, if the autoregressive process began in the recent past, the stationarity constraint = (1 − 2 ) −1 may be poorly approximated. Moreover, Baltagi et al (1992) point out that the unknown starting date of the AR(1) process translates into an "e ective" initial variance 2 * = 2 , which can be estimated with panel data.…”

Section: Basic Experimental Designmentioning

confidence: 99%

“…For = 0, I set = 1. The values of the independent variable x it were generated as in Nerlove (1971); see also Baltagi et al (1992). In particular, it s were generated as independent pseudorandom numbers uniformly distributed over the interval (−1=2; 1=2), and then I set x it = 0:1t + 0:5x i; t−1 + it (t = 1; : : : ; T; i = 1; : : : ; N ); where x i0 = 5 + 10 i0 (i = 1; : : : ; N ).…”

Section: Basic Experimental Designmentioning

confidence: 99%

See 2 more Smart Citations

Estimation of a generalized random-effects model: some ECME algorithms and Monte Carlo evidence

Phillips

2004

Journal of Economic Dynamics and Control

View full text Add to dashboard Cite

Section: Monte Carlo Results For Normal Componentssupporting

confidence: 79%

Section: Basic Experimental Designmentioning

confidence: 99%

Section: Basic Experimental Designmentioning

confidence: 99%

See 1 more Smart Citation

Estimation of a generalized random-effects model: some ECME algorithms and Monte Carlo evidence

Phillips

2004

Journal of Economic Dynamics and Control

View full text Add to dashboard Cite

“…Furthermore, a generalized Chow test is applied to test for poolability of the data and Hausman test is applied to test for model specification. Some of the applications of this technique in the past include Wolpin (1980), Griffin (1982), and more recently Baltagi et al (1992) and Larson and Watters (1993). Table 1 reports the regression results for the beef model.…”

Section: Model Estimation and Empirical Resultsmentioning

confidence: 97%

Exchange rate pass-through, price discrimination, and US meat export prices

2003

View full text Add to dashboard Cite

Exports are becoming increasingly important for US livestock and poultry producers. Consequently, meat industry participants are concerned about the potential impacts of variations in relative currency values. These effects are considered by quantifying the impacts of relative exchange rates on US beef, pork and poultry export prices. In addition, the impacts of GATT and NAFTA agreements on exchange rate pass-through are considered. The results indicate incomplete exchange rate pass-through occurs for several countries. Trade liberalization under GATT has positively influenced US beef and poultry export prices.

show abstract

“…In order to correct for first‐order serial correlation, when the Durbin Watson statistics were not at satisfactory levels, an AR(1) term was added to the regressions in Tables 3 and 4. According to Park et al (2003) and Baltagi and Chang (1992), this estimation method adjusts for first‐order serial correlation and in the case of the particular regressions in question turned Durbin Watson statistics to values around 2. When Durbin Watson statistics are not reported, this implies the absence of values for them that would be problematic.…”

mentioning

confidence: 99%

Does the Capitalization of Development Costs Improve Analyst Forecast Accuracy? Evidence from the UK

Anagnostopoulou¹

2009

Financ Manag Account

View full text Add to dashboard Cite

It has been documented that investments in Research and Development (R&D) are associated with increased errors and inaccuracy in earnings forecasts made by financial analysts. These deficiencies have been generally attributed to information complexity and the uncertainty of the future benefits of R&D. This paper examines whether the capitalization of development costs can reduce analyst uncertainty about the future economic outcome of R&D investments, provide outsiders with a better matching of future R&D-related revenues and costs, and therefore promote accuracy in analyst forecasts. UK data is used, because accounting rules in the United Kingdom permitted firms to conditionally capitalize development costs even before the introduction of the International Financial Reporting Standards. The choice to expense R&D rather than conditionally capitalize development costs is found to relate positively to signed analyst forecast errors. This finding is robust to controlling for the influence of other factors that may affect errors, as well as for the influence of R&D investments on forecast errors. The decision to capitalize versus expense is not observed to have a significant influence on analyst forecast revisions. The findings are interpreted as evidence that the choice to capitalize as opposed to expense may help to reduce deficiencies in analyst forecasts; hence, is informative for users of financial statements. Increased informativeness is expected to have repercussions for the effectiveness with which analysts produce earnings forecasts, and, as a result, market efficiency.

show abstract

Monte Carlo evidence on panel data regressions with AR(1) disturbances and an arbitrary variance on the initial observations

Cited by 14 publications

References 13 publications

Estimation of a generalized random-effects model: some ECME algorithms and Monte Carlo evidence

Estimation of a generalized random-effects model: some ECME algorithms and Monte Carlo evidence

Exchange rate pass-through, price discrimination, and US meat export prices

Does the Capitalization of Development Costs Improve Analyst Forecast Accuracy? Evidence from the UK

Contact Info

Product

Resources

About