Analisis Heterogenitas Spasial Faktor Determinan Popularitas TCG Weiss Schwarz di Jepang Menggunakan Pendekatan Geographically Weighted Poisson Regression(GWPR)

Maulana, Muhammad Faiz (2026) Analisis Heterogenitas Spasial Faktor Determinan Popularitas TCG Weiss Schwarz di Jepang Menggunakan Pendekatan Geographically Weighted Poisson Regression(GWPR). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003221188-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Industri hiburan pop culture di Jepang, khususnya Trading Card Game (TCG) Weiss Schwarz, menunjukkan perkembangan pesat. Keberlangsungan ekosistem fisik ini sangat bergantung pada interaksi sosial komunitas tatap muka dan ketersediaan Local Game Store (LGS). Kondisi ekonomi yang beragam di 47 prefektur Jepang menciptakan heterogenitas spasial terhadap tingkat popularitas permainan ini, yang diukur melalui akumulasi jumlah deck pada turnamen resmi. Pemodelan data cacah konvensional menggunakan Regresi Poisson sering kali mengabaikan variasi spasial yang mengakibatkan estimasi parameter menjadi bias. Penelitian ini memodelkan faktor determinan popularitas Weiss Schwarz dengan mengakomodasi efek heterogenitas spasial menggunakan Geographically Weighted Poisson Regression (GWPR) dengan fungsi pembobot Adaptive Bisquare Kernel. Hasil penelitian membuktikan terjadinya pelanggaran asumsi ekuidispersi pada model Regresi Poisson dengan rasio dispersi sebesar 58,81, yang mengindikasikan adanya pengaruh kewilayahan yang kuat. Kinerja model GWPR terbukti lebih superior dan akurat dalam menyerap efek spasial tersebut, ditandai dengan penurunan nilai Deviance dari 2587,8 menjadi 1676,028 dan penyusutan AICc dari 2594,321 menjadi 1714,224. Model GWPR mampu menjelaskan 73,87% variabilitas data popularitas TCG Weiss Schwarz. Pemetaan signifikansi parameter lokal membagi wilayah Jepang ke dalam dua tipologi wilayah dengan karakteristik yang berbeda. Temuan ini membuktikan bahwa strategi pemasaran tidak dapat diseragamkan secara nasional, melainkan harus dikalibrasi secara spesifik mengikuti struktur sensitivitas ekonomi serta ketersediaan infrastruktur hobi di masing-masing wilayah.
======================================================================================================================================
The pop culture entertainment industry in Japan, particularly the Trading Card Game (TCG) Weiss Schwarz, has shown rapid development. The sustainability of this physical ecosystem heavily relies on face-to-face community social interactions and the availability of Local Game Stores (LGS). Diverse economic conditions across Japan's 47 prefectures create spatial heterogeneity regarding the popularity level of this game, measured through the accumulation of decks in official tournaments. Conventional count data modeling using Poisson Regression often ignores spatial variation, resulting in biased parameter estimations. This study models the determinant factors of Weiss Schwarz's popularity by accommodating spatial heterogeneity effects using Geographically Weighted Poisson Regression (GWPR) with an Adaptive Bisquare Kernel weighting function. The results demonstrate a violation of the equidispersion assumption in the Poisson Regression model with a dispersion ratio of 58.81, indicating a strong regional influence. The performance of the GWPR model proves to be superior and more accurate in absorbing these spatial effects, marked by a decrease in Deviance from 2,587.8 to 1,676.028 and a reduction in AICc from 2,594.321 to 1,714.224. The GWPR model is capable of explaining 73.87% of the variability in the popularity data of the Weiss Schwarz TCG. The mapping of local parameter significance divides the Japanese territory into two regional typologies with distinct characteristics. These findings prove that marketing strategies cannot be standardized nationally; instead, they must be specifically calibrated to follow the structure of economic sensitivity and the availability of hobby infrastructure in each region.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Geographically Weighted Poisson Regression, Heterogenitas Spasial, Jepang, Regresi Poisson, Trading Card Game, Geographically Weighted Poisson Regression, Japan, Spatial Heterogenity, Poisson Regression, Trading Card Game
Subjects:	Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Faiz Maulana
Date Deposited:	03 Aug 2026 04:07
Last Modified:	03 Aug 2026 04:07
URI:	http://repository.its.ac.id/id/eprint/139919

Actions (login required)

View Item