Estimation of regression coefficients after a preliminary test for homoscedasticity

When heteroscedasticity of the variances of disturbances in a regression model is suspected, we perform a preliminary test for homoscedasticity prior to estimation of regression coefficients. According to the result of the pre-test, we use either the ordinary least squares estimator or the two-stage Aitken estimator (2SAE). In this paper, using orthonormal regressors, we derive the mean square error (MSE) of the pre-test estimator and show that the 2SAE is inadmissible when the MSE is used as a criterion. Further, we seek the optimal critical value of the pre-test in the sense of minimizing the average relative risk which is based on the MSE.     

Consumer Attitude towards Organic Labeling Schemes in Japan

ABSTRACT

Japan is the most lucrative organic market in the Asian region and is projected to account for about 80% of the Asian market revenues by 2009 (Kuhlmann and Jones, 2006). This study identifies Japanese consumer preference and interest in two major labeling information- source of organic certification/traceability and country of origin for organic food products. Stated preference method (SPM) based survey data was collected from two major cities in Japan in order to develop Japanese consumers' organic food choice model. Findings show that Japanese consumers are willing to pay 10% price premium for the organic food products compared to conventional products that have no specific labeling, suggesting a significant profit margin to be possibly captured by the organic producers and marketers. The Japanese consumers have perceived value of the JAS label in their stated preference and showed willingness to pay a price premium for this quality certification system, which validates the importance of government-regulated mandatory labeling. Regarding the country origin labeling issue, the Japanese consumers showed clear preference for the domestic organic products to the imported products, while showing no preference for any particular country origin for imported organic products. For private-voluntary organic labeling, the consumer organization-led organic label is considered to be more reliable source than the retailer-guaranteed organic label.     

Currency Substitution in Laos

This study investigates the determinants of currency substitution in Laos based on the function of money demand. The empirical model is estimated using the autoregressive distributed lag approach to cointegration for the period 1993–2012. Our estimation results indicate that the interest rate differential is a significant currency substitution determinant in the Laos economy. Moreover, there is evidence supporting the existence of a ratchet effect in the currency allocation of deposits, implying that particularly strong policies should be pursued over an extended period of time in order to convince depositors to switch back to kip‐denominated assets.     

A switching regression model with different change-points for individual coefficients and its application to the energy demand equations for Japan

In this paper, we set up a switching regression model in which individual coefficients are allowed to shift at different change-points. We also apply it to the energy demand equations and examine structural change in the demands for total fuel oil and for light oil and kerosene at the second oil crisis. It is shown that assuming the different change-points for individual coefficients yields more Plausible results than assuming the same change-point for all coefficients.We thank Professor S. Katayama for his help in another related project.     

Pre-testing on part of the data

Reserving part of a data set for final estimation, using the other part for pre-testing in a regression model is not good practice from a risk function point of view, even though such procedure has been advocated and practised by some econometricians and statisticians. Comparing quadratic risk functions for estimators based on pre-tests on part of the data with estimators based on pre-tests on all the data shows that neither dominate. However, the part sample estimator is decidedly inferior in average relative risk.     

Estimation of variance after a preliminary test of homogeneity and optimal levels of significance for the pre-test

The question of whether to pool two samples in variance estimation is often decided via a preliminary F test. In this paper we show that the optimal pre-test F value is unity for a one- sided alternative, where the objective function is to minimize average relative risk. The outcome is independent of numbers of degrees of freedom in each sample. Optimal significance levels vary somewhat but are close to $\text{1}\text{2}$ for most d.f. and equal to $\text{1}\text{2}$ when numerator and denominator d.f. are equal. The results also apply to regression variance estimation across two data regimes.