Seasonal changes refer to fluctuations caused by recurrent factors, for which non-seasonally adjusted data are unable to reflect the exact changes and direction. Therefore, data users should take into consideration the effect of seasonal factors when carrying out comparisons using non-adjusted data series. The more common seasonal factors include: public holidays (Lunar New Year, Labour Day and National Day with rising number of tourists), specific periods (tax season, commencement of new school year), weather (impact of weather changes on specific industries) and general expectation (production of goods during Christmas season may be affected by high expectation of retail sales).