A Study on the Disaggregation Method of Time Series Data

시계열 자료의 분할에 관한 사례 연구

  • 문승호 (부산외국어대학교 데이터경영학과) ;
  • 이정형 (동아대학교 경영정보학과)
  • Received : 2013.03.25
  • Accepted : 2014.06.20
  • Published : 2014.06.28


When we collect marketing data, we can only obtain the bimonthly or quarterly data but the monthly data be available. If we evaluate or predict monthly market condition or establish monthly marketing strategies, we need to disaggregate these bimonthly or quarterly data to the monthly data. In this paper, for bimonthly or quarterly data, we introduce some methods of disaggregation to monthly data. These disaggregation methods include the simple average method, the growth rate method, the weighting method by the judgment of experts, and variable decomposition method using 12 month moving cumulative sum. In this paper, we applied variable decomposition method to disaggregate for bimonthly data of sum of electronics sales in a European country. We, also, introduce how to use this method to predict the future data.


monthly data;bimonthly data;disaggregation;variable decomposition;prediction


  1. Chaiy, S. (1998), Marketing, 126-128, Hakhyunsa, (in Korean).
  2. Cho, S, Sohn, Y. S. (2011), Time Series Analysis, 3/e, Yulgok Books, (in Korean).
  3. Chow, G. C., Lin, A. (1971). Best Linear Unbiased Interpolation, Distribution and Extrapolation of Time Series by Related Series. Review of Economics and Statististics, 53, 372-375.
  4. Denton, F. T. (1971). Adjustment of monthly or quarterly series to annual totals: an approach Based on quadratic minimization. Journal of American Statistical Association, 66, 99-102.
  5. Lee, H. (1987), Introduction to Management, 394-398, Hyungseol Publishing, (in Korean).
  6. Lee, J., Cho., S. (2001). Forecast of Foreign Tourist Using Time Series model, Journal of the Korean Data Analysis Society, 3(1), 73-86. (in Korean).