No abstract

No abstract

Spectral analysis is applied to weekly supply and price series for bananas traded on the Sydney wholesale market. Many interesting features are revealed such as the pronounced seasonal component and the two week lag structure between the supply and price series. These results are discussed in relation to the known behaviour of the market and some implications for marketing decisions are tentatively drawn. Zn troductionData employed for empirical studies in agricultural marketing are usually available as series recorded at discrete equally-spaced time intervals. As a preliminary step in the analysis of such data a researcher may wish to study its characteristics in some detail in order to describe more precisely the salient features of the underlying mechanism.Various methods such as the periodogram analysis [3] and the correlogram have been used for this purpose in the past, but the spectral technique is rapidly gaining prominence in modern time series analysis as a means of detecting important periodic components in data sets. The technique, originally developed and used in the physical and earth sciences, has recently been employed with economic data as a means of investigating measurable periodicities and also quantifying relationships between pairs of time series [l], [4]. This paper presents a brief description of the supply and price movements on the Sydney Wholesale Banana Market and then discusses an application of the technique using these observed series.

No abstract

No abstract