• The Korean Journal of Applied Statistics
• /
• v.33 no.1
• /
• pp.25-46
• /
• 2020

This paper presents ordinal probit semiparametric regression models using Bayesian Spectral Analysis Regression (BSAR) method. Ordinal probit regression is a way of modeling ordinal responses - usually more than two categories - by connecting the probability of falling into each category explained by a combination of available covariates using a probit (an inverse function of normal cumulative distribution function) link. The Bayesian probit model facilitates posterior sampling by bringing a latent variable following normal distribution, therefore, the responses are categorized by the cut-off points according to values of latent variables. In this paper, we extend the latent variable approach to a semiparametric model for the Bayesian ordinal probit regression with nonparametric functions using a spectral representation of Gaussian processes based BSAR method. The latent variable is decomposed into a parametric component and a nonparametric component with or without a shape constraint for modeling ordinal responses and predicting outcomes more flexibly. We illustrate the proposed methods with simulation studies in comparison with existing methods and real data analysis applied to a Korean National Health and Nutrition Examination Survey (KNHANES) 2016 for investigating nonparametric relationship between smoking behavior and coffee intake.

• Journal of the Korean Society of Food Culture
• /
• v.22 no.6
• /
• pp.794-800
• /
• 2007

The purpose of this study was to measure the tourists' preference for alternative restaurants with different combinations of 4 attribute levels: origin description, food type, price and service guarantee. A total of 210 questionnaires were completed from tourists who visited Kwangyang, Suncheon and Yeosu during Jan. 2 - Jan. 15, 2007. Conjoint experiment method was used to develop hypothetical restaurants. Ordinal probit model was used to measure the effects of attribute levels on the tourists' preference. Results of the study demonstrated that the ordinal probit model analysis result for the data indicated excellent model fit. The effects of attribute levels (origin description, traditional food, fusion food, price, service guarantee) on the tourists' preference were statistically significant. As expected, estimates of marginal willingness to pay for origin description(3.063), food type(2.349), and service guarantee(2.356) were statistically significant. Moreover, tourists were more willing to pay for origin description than other attribute levels. Tourists also considered the origin description as the very important attribute. In conclusion, based on conjoint analysis, a model was proposed of marginal willingness to pay of attribute levels. It should be noted that the original model was modified and should, preferably, be validated in future research.

• Journal of the East Asian Society of Dietary Life
• /
• v.18 no.2
• /
• pp.165-171
• /
• 2008

The purpose of this study was to measure tourists' preference for alternative restaurants with different combinations of attribute levels: grown area logo, origin description, traditional food, fusion food, national food, and price. A total of 210 questionnaires were completed. A conjoint experimental method was used to develop hypothetical restaurants, and an ordinal probit model was used to measure the effects of the attribute levels on tourists' preference. The ordinal probit model analysis results for the data indicated an excellent model fit. The effects of the attribute levels on tourists' preferences were statistically significant. As expected, estimates of the marginal willingness to pay were statistically significant Moreover, the tourists were more willing to pay for grown area logo as compared to the other attribute levels. The tourists also considered the grown area logo as a very important attribute. Withe regard to developing and testing conjoint models in the design of choice experiments involving multifactor alternatives, this study may approach a deeper understanding of the conjoint experiment. Greater understanding of the conjoint experiment can improve the managerial diagnoses of the problems as well as the opportunities for different marketing strategies including local branding programs and menu development and marketing communications.

• Culinary science and hospitality research
• /
• v.14 no.2
• /
• pp.215-224
• /
• 2008

The objective for this research was to clarify the preference for alternative restaurants with different combinations of factor levels: local specialty food, non-local specialty food, very attentive service, moderately attentive service, not attentive service, traditional decoration, modern decoration, \10,000, \15,000, and \20,000. Total 230 copies of questionnaire were completed. Conjoint experiment method was used to develop full restaurant profiles. Ordinal probit model was used to measure the effects of factor levels on the preference. Results of the study demonstrated that the ordinal probit model analysis result for the data also indicated a good model fit. The effects of factor levels on the preference were statistically significant. As expected, the estimates of implicit price to pay were statistically significant. Moreover, the customers were more willing to pay for local specialty than other factor levels. The customers also considered the food factor as a very important factor. This research suggested that the customer's decision-making process for restaurants was best modeled as a conjoint experiment method that combines various factor levels. And it showed the results could be used as good data for understanding the relationships between the factors and preference in choosing food and restaurants in tourist areas.

• Communications for Statistical Applications and Methods
• /
• v.30 no.1
• /
• pp.49-63
• /
• 2023

A lot of studies on the summary measures of predictive strength of categorical response models consider the likelihood ratio index (LRI), also known as the McFadden-R², a better option than many other measures. We propose a simple modification of the LRI that adjusts for the effect of the number of response categories on the measure and that also rescales its values, mimicking an underlying latent measure. The modified measure is applicable to both binary and ordinal response models fitted by maximum likelihood. Results from simulation studies and a real data example on the olfactory perception of boar taint show that the proposed measure outperforms most of the widely used goodness-of-fit measures for binary and ordinal models. The proposed R² interestingly proves quite invariant to an increasing number of response categories of an ordinal model.

• The Korean Journal of Applied Statistics
• /
• v.25 no.2
• /
• pp.363-376
• /
• 2012

Excessive zeroes are often observed in ordinal categorical response variables. An ordinary ordered Probit model is not appropriate for zero-inflated data especially when there are many different sources of generating 0 observations. In this paper, we apply a two-stage zero-inflated ordered Probit (ZIOP) model which incorporate the zero-flated nature of data, propose a Bayesian analysis of a ZIOP model, and apply the method to alcohol consumption data collected by the National Bureau of Statistics, Korea. In the first stage of a ZIOP model, a Probit model is introduced to divide the non-drinkers into genuine non-drinkers who do not participate in drinking due to personal beliefs or permanent health problems and potential drinkers who did not drink at the time of the survey but have the potential to become drinkers. In the second stage, an ordered probit model is applied to drinkers that consists of zero-consumption potential drinkers and positive consumption drinkers. The analysis results show that about 30% of non-drinkers are genuine non-drinkers and hence the Korean alcohol consumption data has the feature of zero-inflated data. A study on the marginal effect of each explanatory variable shows that certain explanatory variables have effects on the genuine non-drinkers and potential drinkers in opposite directions, which may not be detected by an ordered Probit model.

• Culinary science and hospitality research
• /
• v.14 no.1
• /
• pp.1-10
• /
• 2008

The purpose of this study was to measure the estimated coefficient and marginal willingness to pay of attribute level: origin logo, origin description, traditional food, fusion food, service guarantee, and price, which influence tourist preference for restaurants. Also, the study identified the attribute which confers the highest importance to tourists. Conjoint experiment and the ordinal probit model were used for this study. A total of 210 surveys were conducted by tourists. The findings from this study were as follows. First, the effects of all attributes on tourist preference for restaurants were statistically significant. Second, tourists regarded an origin logo as the very important attribute, and were more willing to pay for the case where the menu contained origin logo. These findings suggested new marketing opportunities for restaurant managers. Although this study provides some evidence on the value of the local product brand to tourists, a similar measure has not been developed for local residents. This is an area in need of future research.