In this paper an appropriate model of the seasonal pattern in agricultural data is proposed, which takes the specific nature of such a pattern into account. The methodological proposal is based on evolving splines that are shown to be a tool capable of modeling seasonal variations in which either the period or the magnitude of the seasonal fluctuations do not remain the same over time. This proposal is applied to capture the movements in a weekly tomato export series and the analysis is carried out inside the frame delimited by the structural approach to time series.

Agricultural time series data recorded at intervals shorter than a year are increasingly used in the economic literature. As Engle (2000) pointed out, much of the movement to higher frequency analysis was a consequence of the availability of higher frequency measurements and it is supposed that this will continue and that ever increasing frequencies of observations will be used. In fact, additional research on agricultural economics applied to monthly, weekly, daily or hourly series or even to data recorded at shorter intervals, is usually needed. Decision-makers in agricultural economics should systematically use seasonal information as a way to diminish the risks associated with some uncertainties affecting the business activity. The detection of any identifiable and regular seasonal patterns in the temporal behavior of agricultural prices could be used as a basis to plan on-farm temporal supply. Monthly or weekly data are convenient in order to observe clearly the price transmission mechanism in agricultural markets (Miller and Hayenga, 2001; Sanjuan and Dawson, 2003) or the term structures of futures prices over time (Sorensen, 2002) and also to analyze the retailers strategies (Richards and Patterson, 2005). The time to flowering of crops depends on hourly solar radiation and temperature and daily rainfall data are also relevant. In general, there are important seasonal climate effects on agricultural exports for supplier countries (Hill et al., 2004). On the other hand, some papers show that the daily wage variation is an important aspect of labor-market adjustment (Jarvis and Vera-Toscano, 2004). In other papers, daily futures prices are used to analyze the structure of market reaction to news (Rucker et al., 2005). In summary, better knowledge of seasonal variability is valuable for agricultural decision-making at the farm, marketing or policy level.

Heterogeneous Seasonal Patterns in Agricultural Data and Evolving Splines,agricultural, seasonal, prices, economics, frequency, intervals, pattern, proposal, recorded, analyze, temporal, climate, consequence, countries, delimited, detection, economic, evolving, flowering, fluctuations, frequencies, identifiable, increasingly, information, knowledge, literature, magnitude, measurements, mechanism, methodological, modeling, movements, observations, appropriate, onfarm, Patterson, associated