5.2.2 Graphical analysis of price trends
Now that we know what data are available, and understand the general contextual
information, it is time to finally review some price data. The general approach that is
suggested in this manual is to review the historical price data using graphical analysis.
There are many methods to observe historical price trends, ranging from simple to
complex. For the purposes of price analysis in early warning simpler methods are
appropriate. Graphical analysis involves plotting (graphing) the price data for a
particular commodity market combination over time and explaining the observed pattern.
This approach is suggested since it is an easy way to catch major trends and anomalies in
During this process the contextual information about agricultural production, consumption,
and marketing generated earlier will be critical to explain the behavior of prices and
markets. Through this process you will be able to understand how this market usually
behaves after different production outcomes (average, above average, and below average),
when structural changes occurred in the economy, or other shocks. Only when one
understands the behavior of a market over time can one begin to explain anomalous or odd
behavior. It is the identification of the anomalous behavior that is important in early
warning, especially the identification of a potentially problematic behavior that will
adversely affect household food security. As will be discussed later, the reporting of the
anomalous behavior requires a reference period for comparison, which is what the review of
the historical data will do.
At least two years of data is necessary to see historical trends. To have a clearer picture of
historical price trends one would ideally need five or more years. If only one or two years
of price data are available for a given market or commodity, more care should be made to
interpret the behavior of current prices in those markets.
There are four steps that are important to do when reviewing historical price trends: 1)
understanding seasonal patterns, 2) characterizing price relationships across commodities in
the same market, 3) characterizing price relationships for a single commodity across
markets, and 4) understanding relative prices. For each step of analysis, an explanation will
be provided as to the purpose of the analysis and how the results should be interpreted.
During the analysis of historical price data there are various aspects of price behavior that
should be evaluated. The aspects to evaluate are the shape, rate of change, and peak of the
historical price curves. This information should be correlated with the contextual
information about production, consumption, and marketing patterns. Any anomalies in the
historical times series data should be investigated. The first step to checking on these
anomalies is to crosscheck with other data/information you have or know about that area. If
necessary, more intensive investigation should be conducted with either key informant
interviews or field trips.