This paper adopts Hsiao‘s panel data model technique. In this study, random effect model in the one-way error component regression model has been determined. Analyzing Seoul upscale hotels, over a 36 quarter period (1997Q1~2005Q4), this study provides empirical evidences of significance positive relationships between the number of arrivals in Korea, exchange rate, seasonality of 2 quarter, 4 quarter and room demand of foreign user. However, this study provides empirical evidences of significance negative relationships between average room rate, consumer price index, severe acute respiratory syndrom, pathogenic avian influenza and room demand of foreign user. Implications of these findings are discussed for hospitality context research and practice.핵심용어(Key words): 객실수요 결정요인(Determinants of room demand), 패널데이터모형(Panel data model), 임의효과모형(Random-effects model), 고정효과모형(Fixed-effects model), 개별특성효과(Individual-specificeffects), 시간특성효과(Time-specific effects)