Papers - Hoshino, Takahiro

Doubly Robust-type Estimation of Population Moments or Parameters in Biased Sampling

Takahiro Hoshino and Yuya Shimizu

Stat in press 2019

Research paper (scientific journal), Single Work, Accepted

Bayesian Data Combination Approach for Repeated Durations under Unobserved Missing Indicators

Ryosuke Igari and Takahiro Hoshino

Computational Statistics & Data Analysis (Elsevier B.V.)  126   150 - 166 2018.10

Research paper (scientific journal), Single Work, Accepted,  ISSN  01679473

　View Summary

Intermittent missingness in repeated duration analysis is common in applied studies, but has not been rigorously considered in statistics. Under intermittent missingness, whether any missing events exist between two observed events is unknown. In other words, the missing indicators are never observed. Thus, if there exist any missing events between two observed events, researchers observe only the cumulative duration of the two or more events. A quasi-Bayes estimation method that utilizes population-level information is used to appropriately estimate the parameters under unobserved intermittent missingness. The proposed model consists of the following: (1) a latent variable model, (2) a latent missing indicator model separating the true and composite durations, (3) mixtures of duration models, and (4) moment restriction from population-level information to deal with nonignorable intermittent missingness. A new estimation procedure is used to simultaneously combine likelihood and the objective function of GMM with the latent variables
this is called Bayesian data combination. The model is applied to the interpurchase duration in database marketing using the purchase history data of Japan
these data capture the purchase incidences and stores.

• Access to Document (DOI)

中高生におけるインターネット上のいじめ被害と加害の関連性：大規模横断調査による検討

浦野由平・大賀真伊・滝沢龍・星野崇宏・下山晴彦

臨床心理学 18 ( 4 ) 486 - 491 2018.07

Research paper (scientific journal), Single Work, Accepted

Effectiveness of Influenza Vaccination for Children in Japan: Four-year Observational Study Using a Large-scale Claims Database

HOSHINO Takahiro

Vaccine (Vaccine)  36 ( 20 ) 2809 - 2815 2018.05

Research paper (scientific journal), Single Work, Accepted,  ISSN  0264410X

　View Summary

© 2018 Elsevier Ltd Background: To date, few large-scale comparative effectiveness studies of influenza vaccination have been conducted in Japan, since marketing authorization for influenza vaccines in Japan has been granted based only on the results of seroconversion and safety in small-sized populations in clinical trial phases not on the vaccine effectiveness. We evaluated the clinical effectiveness of influenza vaccination for children aged 1–15 years in Japan throughout four influenza seasons from 2010 to 2014 in the real world setting. Methods: We conducted a cohort study using a large-scale claims database for employee health care insurance plans covering more than 3 million people, including enrollees and their dependents. Vaccination status was identified using plan records for the influenza vaccination subsidies. The effectiveness of influenza vaccination in preventing influenza and its complications was evaluated. To control confounding related to influenza vaccination, odds ratios (OR) were calculated by applying a doubly robust method using the propensity score for vaccination. Results: Total study population throughout the four consecutive influenza seasons was over 116,000. Vaccination rate was higher in younger children and in the recent influenza seasons. Throughout the four seasons, the estimated ORs for influenza onset were statistically significant and ranged from 0.797 to 0.894 after doubly robust adjustment. On age stratification, significant ORs were observed in younger children. Additionally, ORs for influenza complication outcomes, such as pneumonia, hospitalization with influenza and respiratory tract diseases, were significantly reduced, except for hospitalization with influenza in the 2010/2011 and 2012/2013 seasons. Conclusions: We confirmed the clinical effectiveness of influenza vaccination in children aged 1–15 years from the 2010/2011 to 2013/2014 influenza seasons. Influenza vaccine significantly prevented the onset of influenza and was effective in reducing its secondary complications.

• Access to Document (DOI)

大規模調査データによる消費者の保険加入行動の理解(1)：属性やリスク資産保有との関係

竹内　真登・星野　崇宏

損害保険研究 79 ( 4 ) 1 - 38 2018.02

Research paper (scientific journal), Single Work

Inadvertent Learning on a Portal Site: A Longitudinal Field Experiment

Tetsuro Kobayashi, Takahiro Hoshino and Takashisa Suzuki

Communications Research in press 2018

Research paper (scientific journal), Joint Work, Accepted

• Access to Document (DOI)

A Survival Analysis Incorporating Auxiliary Information by Bayesian Generalized Method of Moment: Application to Interpurchase-timing Modeling

Ryosuke Igari and Takahiro Hoshino

Journal of the Japanese Society of Computational Statistics 30 ( 1 ) 27 - 44 2017.12

Research paper (scientific journal), Single Work, Accepted

• Access to Document (DOI)

企業ブランドイメージと広告費、財務指標の 関連性に関する実証分析

加藤諒・星野崇宏

日経広告研究所報 294   18 - 25 2017.09

Single Work

An Out-of-Pocket Cost Removal Intervention on Fecal Occult Blood Test Attendance

Takahiro Tabuchi, Hiroshi Murayama, Takahiro Hoshino, and Tomio Nakayama

American Journal of Preventive Medicine (American Journal of Preventive Medicine)  53 ( 2 ) e51 - e62 2017.08

Research paper (scientific journal), Single Work, Accepted,  ISSN  07493797

　View Summary

© 2017 American Journal of Preventive Medicine Introduction To date, no comparative study has assessed the impact of a cost-removal intervention on fecal occult blood testing (FOBT). In 2012, the Japanese government introduced a nationwide project to remove out-of-pocket costs for FOBT. The study objective was to evaluate the differential impact of the intervention on FOBT attendance in the total population and various subgroups. Methods This study analyzed 309,103 people in national, repeated cross-sectional studies, observed pre- and post-intervention (2010 and 2013), using covariate-adjusted difference-in-differences estimates to compare intervention and no-intervention groups. The outcome measure was uptake of FOBT attendance resulting from the intervention. Stratified analyses were conducted according to sociodemographic and health-related characteristics. Results The intervention was associated with significantly positive uptake of FOBT in both genders, but the impact was greater in women than men: 6.7% (95% CI=5.2, 8.1) for women and 2.7% (95% CI=1.1, 4.3) for men in the covariate-adjusted models. Post-intervention, attendance increased in almost all subgroups in women. However, among men, some socially advantaged subgroups, such as high expenditure, high education, and public officers, showed no effect. Some subgroups such as current smokers and less than high school education were identified as hard-to-reach populations that may be less sensitive to the intervention, irrespective of gender. Conclusions This is the first comparative study of cost-removal intervention for uptake of FOBT. The intervention may increase FOBT attendance. However, the size of the effect is not great, especially in men, and differential effects occurred across subgroups including gender and socioeconomic differences.

• Access to Document (DOI)

The impact of product based reward promotions: Which sales strategies are more effective for retail companies, price discounts or point incentives?

NAKAGAWA Hiromichi and HOSHINO Takahiro

Journal of Marketing & Distribution 20 ( 2 ) 1 - 15 2017.07

Research paper (scientific journal), Single Work, Accepted

Respondents with low motivation tend to choose middle category: Survey questions on happiness in Japan

Shinya Masuda, Takayuki Sakagami, Hideaki Kawabata, Nobuhiko Kijima and Takahiro Hoshino

Baheviormetrika 44 ( 2 ) 593 - 605 2017.06

Research paper (scientific journal), Joint Work, Accepted

• Access to Document (DOI)

Researchers’ Choice of Number and Range of Levels in Experiments Affects the Resultant Variance-Accounted-For Effect Size

Kenshuke Okada, Takahiro Hoshino

Psychonomic Bulletin & Review (Psychonomic Bulletin and Review)  24 ( 2 ) 607 - 616 2017.04

Joint Work, Accepted,  ISSN  10699384

　View Summary

© 2016, Psychonomic Society, Inc. In psychology, the reporting of variance-accounted-for effect size indices has been recommended and widely accepted through the movement away from null hypothesis significance testing. However, most researchers have paid insufficient attention to the fact that effect sizes depend on the choice of the number of levels and their ranges in experiments. Moreover, the functional form of how and how much this choice affects the resultant effect size has not thus far been studied. We show that the relationship between the population effect size and number and range of levels is given as an explicit function under reasonable assumptions. Counterintuitively, it is found that researchers may affect the resultant effect size to be either double or half simply by suitably choosing the number of levels and their ranges. Through a simulation study, we confirm that this relation also applies to sample effect size indices in much the same way. Therefore, the variance-accounted-for effect size would be substantially affected by the basic research design such as the number of levels. Simple cross-study comparisons and a meta-analysis of variance-accounted-for effect sizes would generally be irrational unless differences in research designs are explicitly considered.

• Access to Document (DOI)

顧客行動の多様性変数を利用した購買行動の予測

新美　潤一郎・星野　崇宏

人工知能学会論文誌 32 ( 2 ) B-G63_1 - B-G63_9 2017.03

Research paper (scientific journal), Single Work, Accepted

• Access to Document (DOI)

プロセスシミュレーションを伴うコンジョイント測定による購買予測：写真提示を用いた操作と追跡調査による予測精度向上の確認

竹内　真登・星野　崇宏

行動計量学 44 ( 1 ) 45 - 56 2017.03

Research paper (scientific journal), Single Work, Accepted

Predicting purchases with using the variety of customer behaviors: Analysis of the purchase history and the browsing history by deep learning

Niimi J., Hoshino T.

Transactions of the Japanese Society for Artificial Intelligence (Transactions of the Japanese Society for Artificial Intelligence)  32 ( 2 )  2017

ISSN  13460714

　View Summary

© 2017, Japanese Society for Artificial Intelligence. All rights reserved. Nowadays, along with the popularity of E-Commerce, the marketing strategy of retail stores has been more complicated with O2O or Omni-channel. Therefore, Customer Relationship Management (CRM) is one of the important issue for the retail stores. It can be difficult to predict customers future behavior with the simple quantitive information such as purchase frequency since each customers are widely diversified. Although the company can obtain the variety of customers information from their online activity, the use of access history is still limited. In this paper, we defined “the variety of user access patterns” collected from their web browsing history and it shows the patterns they visit the website. Finally, we verified its effectiveness with developing a DNN model to predict customers future behavior.

効果的な効果検証？：非無作為化デザインによる刑事政策の因果効果の推定

高橋哲・只野智弘・星野崇宏

更生保護学研究 (日本更生保護学会) 9   52 - 74 2016.12

Research paper (scientific journal), Single Work, Accepted

Online-Offlineチャネルにおける消費者の購買間隔と購買金額の同時モデリング

猪狩良介・星野　崇宏

オペレーションズ・リサーチ 61 ( 9 ) 589 - 599 2016.09

Single Work, Accepted

商品カテゴリー購買と複数ブランド購買の段階型同時分析モデル

宮崎　慧・星野　崇宏

行動計量学 43 ( 2 ) 167 - 180 2016.09

Single Work, Accepted

ユーザ別アクセス・パターン情報の多様性を用いた顧客行動の予測とモデリング

HOSHINO TAKAHIRO

応用統計学 in press ( 3 ) 121 - 143 2016

Research paper (scientific journal), Joint Work, Accepted

• Access to Document (DOI)

A national survey of psychology education programs and their content in Japan

HOSHINO TAKAHIRO

Japanese Psychological Research (Japanese Psychological Research)  in press ( S1 ) 4 - 18 2016

Research paper (scientific journal), Joint Work, Accepted,  ISSN  00215368

　View Summary

© 2016 Japanese Psychological Association. Published by John Wiley & Sons Australia, Ltd. This study reports the results of a national survey of psychology programs in Japan, conducted by the Committee on the Education and Research of the Japanese Psychological Association, based on 443 responses from program heads of 378 colleges, universities, and graduate schools. We analyzed the data by institution and type of degree program. The results indicated that: (a) undergraduate-level psychological education in Japan is mainly provided by private universities (which account for 71% of all psychology graduates at this level), while Master's (60%) and Ph.D. (83%) programs are mainly provided by national and public universities; (b) the largest groups of faculty members by specialization are in clinical psychology and in developmental and educational psychology; (c) the dominant forms of teaching are lectures and seminars; and (d) most psychology major programs in Japan aim to improve students' academic and generic skills. Finally, we suggest how psychology programs in Japan can reform curriculum and better improve students' academic skills on the basis of their generic skills.

• Access to Document (DOI)