Search results
1 – 10 of 890Benjamin J. Gillen, Matthew Shum and Hyungsik Roger Moon
Structural models of demand founded on the classic work of Berry, Levinsohn, and Pakes (1995) link variation in aggregate market shares for a product to the influence of product…
Abstract
Structural models of demand founded on the classic work of Berry, Levinsohn, and Pakes (1995) link variation in aggregate market shares for a product to the influence of product attributes on heterogeneous consumer tastes. We consider implementing these models in settings with complicated products where consumer preferences for product attributes are sparse, that is, where a small proportion of a high-dimensional product characteristics influence consumer tastes. We propose a multistep estimator to efficiently perform uniform inference. Our estimator employs a penalized pre-estimation model specification stage to consistently estimate nonlinear features of the BLP model. We then perform selection via a Triple-LASSO for explanatory controls, treatment selection controls, and instrument selection. After selecting variables, we use an unpenalized GMM estimator for inference. Monte Carlo simulations verify the performance of these estimators.
Details
Keywords
Xiang Chen, Yaohui Pan and Bin Luo
One challenge for tourism recommendation systems (TRSs) is the long-tail phenomenon of ratings or popularity among tourist products. This paper aims to improve the diversity and…
Abstract
Purpose
One challenge for tourism recommendation systems (TRSs) is the long-tail phenomenon of ratings or popularity among tourist products. This paper aims to improve the diversity and efficiency of TRSs utilizing the power-law distribution of long-tail data.
Design/methodology/approach
Using Sina Weibo check-in data for example, this paper demonstrates that the long-tail phenomenon exists in user travel behaviors and fits the long-tail travel data with power-law distribution. To solve data sparsity in the long-tail part and increase recommendation diversity of TRSs, the paper proposes a collaborative filtering (CF) recommendation algorithm combining with power-law distribution. Furthermore, by combining power-law distribution with locality sensitive hashing (LSH), the paper optimizes user similarity calculation to improve the calculation efficiency of TRSs.
Findings
The comparison experiments show that the proposed algorithm greatly improves the recommendation diversity and calculation efficiency while maintaining high precision and recall of recommendation, providing basis for further dynamic recommendation.
Originality/value
TRSs provide a better solution to the problem of information overload in the tourism field. However, based on the historical travel data over the whole population, most current TRSs tend to recommend hot and similar spots to users, lacking in diversity and failing to provide personalized recommendations. Meanwhile, the large high-dimensional sparse data in online social networks (OSNs) brings huge computational cost when calculating user similarity with traditional CF algorithms. In this paper, by integrating the power-law distribution of travel data and tourism recommendation technology, the authors’ work solves the problem existing in traditional TRSs that recommendation results are overly narrow and lack in serendipity, and provides users with a wider range of choices and hence improves user experience in TRSs. Meanwhile, utilizing locality sensitive hash functions, the authors’ work hashes users from high-dimensional vectors to one-dimensional integers and maps similar users into the same buckets, which realizes fast nearest neighbors search in high-dimensional space and solves the extreme sparsity problem of high dimensional travel data. Furthermore, applying the hashing results to user similarity calculation, the paper greatly reduces computational complexity and improves calculation efficiency of TRSs, which reduces the system load and enables TRSs to provide effective and timely recommendations for users.
Details
Keywords
Single-shot multi-category clothing recognition and retrieval play a crucial role in online searching and offline settlement scenarios. Existing clothing recognition methods based…
Abstract
Purpose
Single-shot multi-category clothing recognition and retrieval play a crucial role in online searching and offline settlement scenarios. Existing clothing recognition methods based on RGBD clothing images often suffer from high-dimensional feature representations, leading to compromised performance and efficiency.
Design/methodology/approach
To address this issue, this paper proposes a novel method called Manifold Embedded Discriminative Feature Selection (MEDFS) to select global and local features, thereby reducing the dimensionality of the feature representation and improving performance. Specifically, by combining three global features and three local features, a low-dimensional embedding is constructed to capture the correlations between features and categories. The MEDFS method designs an optimization framework utilizing manifold mapping and sparse regularization to achieve feature selection. The optimization objective is solved using an alternating iterative strategy, ensuring convergence.
Findings
Empirical studies conducted on a publicly available RGBD clothing image dataset demonstrate that the proposed MEDFS method achieves highly competitive clothing classification performance while maintaining efficiency in clothing recognition and retrieval.
Originality/value
This paper introduces a novel approach for multi-category clothing recognition and retrieval, incorporating the selection of global and local features. The proposed method holds potential for practical applications in real-world clothing scenarios.
Details
Keywords
Roman Liesenfeld, Jean-François Richard and Jan Vogler
We propose a generic algorithm for numerically accurate likelihood evaluation of a broad class of spatial models characterized by a high-dimensional latent Gaussian process and…
Abstract
We propose a generic algorithm for numerically accurate likelihood evaluation of a broad class of spatial models characterized by a high-dimensional latent Gaussian process and non-Gaussian response variables. The class of models under consideration includes specifications for discrete choices, event counts and limited-dependent variables (truncation, censoring, and sample selection) among others. Our algorithm relies upon a novel implementation of efficient importance sampling (EIS) specifically designed to exploit typical sparsity of high-dimensional spatial precision (or covariance) matrices. It is numerically very accurate and computationally feasible even for very high-dimensional latent processes. Thus, maximum likelihood (ML) estimation of high-dimensional non-Gaussian spatial models, hitherto considered to be computationally prohibitive, becomes feasible. We illustrate our approach with ML estimation of a spatial probit for US presidential voting decisions and spatial count data models (Poisson and Negbin) for firm location choices.
Details
Keywords
This paper investigates the capital structure of a large sample of U.S. private firms from 2004 to 2013. There is a considerable heterogeneity in private firm capital structure…
Abstract
This paper investigates the capital structure of a large sample of U.S. private firms from 2004 to 2013. There is a considerable heterogeneity in private firm capital structure not only in terms of the level of leverage but also with regard to the issuance of specific debt instruments. Leverage, debt type usage, and debt specialization are dynamic and strongly related to observable firm characteristics largely in support of contract theory. Unobservable firm and industry characteristics are strong determinants of leverage levels and debt specialization. Macro credit conditions are not related to private firm leverage but are strong determinants of the degree to which firms diversify their debt capital structures.
Details
Keywords
Qiangqiang Zhai, Zhao Liu, Zhouzhou Song and Ping Zhu
Kriging surrogate model has demonstrated a powerful ability to be applied to a variety of engineering challenges by emulating time-consuming simulations. However, when it comes to…
Abstract
Purpose
Kriging surrogate model has demonstrated a powerful ability to be applied to a variety of engineering challenges by emulating time-consuming simulations. However, when it comes to problems with high-dimensional input variables, it may be difficult to obtain a model with high accuracy and efficiency due to the curse of dimensionality. To meet this challenge, an improved high-dimensional Kriging modeling method based on maximal information coefficient (MIC) is developed in this work.
Design/methodology/approach
The hyperparameter domain is first derived and the dataset of hyperparameter and likelihood function is collected by Latin Hypercube Sampling. MIC values are innovatively calculated from the dataset and used as prior knowledge for optimizing hyperparameters. Then, an auxiliary parameter is introduced to establish the relationship between MIC values and hyperparameters. Next, the hyperparameters are obtained by transforming the optimized auxiliary parameter. Finally, to further improve the modeling accuracy, a novel local optimization step is performed to discover more suitable hyperparameters.
Findings
The proposed method is then applied to five representative mathematical functions with dimensions ranging from 20 to 100 and an engineering case with 30 design variables.
Originality/value
The results show that the proposed high-dimensional Kriging modeling method can obtain more accurate results than the other three methods, and it has an acceptable modeling efficiency. Moreover, the proposed method is also suitable for high-dimensional problems with limited sample points.
Details
Keywords
Daejin Kim, Hyoung-Goo Kang, Kyounghun Bae and Seongmin Jeon
To overcome the shortcomings of traditional industry classification systems such as the Standard Industrial Classification Standard Industrial Classification, North American…
Abstract
Purpose
To overcome the shortcomings of traditional industry classification systems such as the Standard Industrial Classification Standard Industrial Classification, North American Industry Classification System North American Industry Classification System, and Global Industry Classification Standard Global Industry Classification Standard, the authors explore industry classifications using machine learning methods as an application of interpretable artificial intelligence (AI).
Design/methodology/approach
The authors propose a text-based industry classification combined with a machine learning technique by extracting distinguishable features from business descriptions in financial reports. The proposed method can reduce the dimensions of word vectors to avoid the curse of dimensionality when measuring the similarities of firms.
Findings
Using the proposed method, the sample firms form clusters of distinctive industries, thus overcoming the limitations of existing classifications. The method also clarifies industry boundaries based on lower-dimensional information. The graphical closeness between industries can reflect the industry-level relationship as well as the closeness between individual firms.
Originality/value
The authors’ work contributes to the industry classification literature by empirically investigating the effectiveness of machine learning methods. The text mining method resolves issues concerning the timeliness of traditional industry classifications by capturing new information in annual reports. In addition, the authors’ approach can solve the computing concerns of high dimensionality.
Details
Keywords
Yujie Zheng and Meiyan Li
Improving the prediction accuracy of design time for complex products is significant for improving the accuracy of product development and control plans. The purpose of this study…
Abstract
Purpose
Improving the prediction accuracy of design time for complex products is significant for improving the accuracy of product development and control plans. The purpose of this study is to propose an intelligent pre-estimation method of design time for complex products based on v-SVM.
Design/methodology/approach
First, an evaluation model for designer knowledge abilities based on v-SVM is built, which considers the fuzziness and dynamics of designer knowledge abilities. Next, a pre-estimation method for the design time of complex products based on v-SVM is built. This method takes into account the impacts of designer knowledge abilities and design task characteristics on the design time. Then, an adaptive genetic algorithm is programmed to optimize the parameters in the evaluation model and the pre-estimation method. Finally, a practical application and comparative analysis of the proposed pre-estimation method is suggested to verify the validity and applicability of this research.
Findings
First, the evaluation of designer knowledge abilities is a prediction problem that is both fuzzy and multivariate time series. Second, the pre-estimation of design time is a problem that is fuzzy and multivariate. Third, the pre-estimation accuracy of the proposed method is higher when compared with traditional methods.
Originality/value
This paper presents an intelligent pre-estimation method of design time for complex products. Unlike previous research, the pre-estimation method takes into account the impacts of both the designer knowledge abilities and the design task characteristics on the design time.
Details
Keywords
A method is proposed for handling multi-attribute judgment problems with a large number of attributes such as mobile phone features. To minimize the complication of…
Abstract
Purpose
A method is proposed for handling multi-attribute judgment problems with a large number of attributes such as mobile phone features. To minimize the complication of multi-attributes and reduce the consumers’ choice task burden, this paper aims to suggest an integrated hierarchical survey design (IHSD) with the Kano model. The author compared the utility of mobile phone’s attributes for each market and for customer segment by analyzing empirical data on wear obtained from six Middle East and African countries, five Asia-Pacific countries and three European countries. Based on an IHSD of 10,200 respondents, brand, camera, memory and LTE (4G) play vital roles in all regions. In contrast, Wi-Fi, file-editor, MMS, LCD size and phone type are displayed as the least important attributes. The results of this study were successfully implemented for product planning, product development and marketing strategy in terms of price setting, features prioritizing and optimal designing for new products in the mobile phone company.
Design/methodology/approach
The first step was to list all possible features with the product planning team, product development team and market research specialists. The second step divided the selected features for designing a mobile phone into subgroups based on their functional characteristics by using the Kano model. The method for classifying features was determined using Kano questionnaire. The third step incorporated a fractional factorial design for the “must-be” choice-based conjoint (CBC) (Oppewal et al., 1994) which includes two factors: whether customers required the “one-dimensional” feature or the “attractive” feature, along with the “must-be” attributes. The consumers who selected the “must-be” features could choose both the “one-dimensional” feature and the “attractive” feature groups or one of the two feature groups in no particular order. Fractional factorial design was applied to both the “one-dimensional” features and the “attractive” features for individual CBCs. Random sequences of the combinations of attribute levels were generated for each of the three types of CBC analyses (“must-be”, “one-dimensional” and “attractive”). At the same time, the fourth step conducted a survey of the individual groups for the conjoint analysis on the functional characteristics of a mobile phone. The analysis of the accumulated data obtained from all the feature groups was completed using conditional logit models as part of the fifth step. In addition, the “must-be” CBC design was linked with the “one-dimensional” and “attractive” CBC designs. The sixth step was to analyze the accumulated results obtained from all the feature groups and estimate the usefulness of each feature’s level in the context of the CBC. Based on the results of the sixth step, the importance and willingness-to-pay of each attribute were estimated in the seventh step.
Findings
Use of the conjoint important score is aimed to expand the market by finding the different consumers’ needs across the regions. In detail, attributes such as “FM Transmitter”, “Touch screen” and “Health (heart rate)” are considered consumers’ new crucial needs in Europe, which would enable the product to superiorly differentiate itself from others to dominate the current market. On the other hand, it is shown that attributes such as “brand”, “mobile TV”, “external memory”, “mobile tracker” and “4G” are more important in Asia-Pacific. Therefore, if mobile manufacturers develop this sector more, it will grant mobile manufacturers the opportunity to lead the market. The only difference of the Middle East and African consumers is that “NFC” has a higher importance while the rest of the needs are very similar to those of Asia-Pacific. Regarding willingness-to-pay (WTP) among countries, the highest scoring utility, besides brand, appeared to be associated with the camera function in all countries. Especially, relatively low utility value was given in Wi-Fi and File-editer, MMS, LCD size and Phone type. In a value-based approach, the price of a product is based on the perceived valuation by the target customers. The research in the field of pricing is of ample importance. This is because price is the only element of the marketing mix that generates income. All other elements, such as advertising and promotion, product development, selling effort, distribution and packaging, involve expenditures (Monroe, 2003). Regarding among regions, the needs for 3G and the internet-related feature (WAP, Wi-Fi, etc.) in the emerging market are low compared to those for 4G and internet-related feature in the mature market. Also, the needs for productivity and advanced features, such as camera and e-mail, are lower in Asia-Pacific than in Europe. It is therefore recommended that manufactures and marketers of mobile phones should consider producing and selling phones with modern technology features that are more durable and of highly quality.
Research limitations/implications
The integrated hierarchical survey by function with the Kano model proves to be a highly useful, efficient and accurate methodology for understanding a consumer mobile phone behavior. Although the proposed method was applied to designs of mobile phones in the emerging and mature markets, its accuracy was not compared with the traditionally used methods such as CBC, adaptive conjoint analysis and hybrid method. This is left for further areas of research.
Practical implications
The results of this research study correspond with previous studies conducted (Pakola et al., 2010; Das, 2012; Malaasi, 2012, 2008; Dziwornu, 2013), which consider the features of mobile phone as a crucial factor in consumer buying decision in all countries. It is significant that this study made huge impact on mobile phone manufacturers in several ways. It has been converted into product development with consumer-oriented approach. The pricing policy has been changed from cost-based pricing into value-based pricing; and marketing strategy has been changed from an unsystematic function into a systematic and consistent one.
Originality/value
The proposed method with the Kano model proved to be a practical and efficient tool for decision-making, as it helped mobile manufacturers to better understand how customers evaluate and perceive quality attributes. The Kano model was used to explain how the quality attributes can be classified into mainly three categories of perceived quality: “must-be”, “one-dimensional” and “attractive”. It has lots of benefits in terms of cost and time reduction and is expected to bring a great effect into the industrial field.
Details