6.2.2 Analysis activities
As an early warning analyst, there are many tasks competing for your time. A plan has to
be developed to facilitate monitoring activities. This set of procedures is required to
analyze price data efficiently. This section details the procedures needed to provide
timely and useful reporting of price information. The relative responsibilities for the staff
within the early warning unit need to be clear to avoid confusion and ensure rapid, high
There are two levels of analysis in monitoring agricultural prices. The first is the regular
plotting and review of the data as they are received from the market information system.
The second activity is the further investigation of where problems seem to be developing.
There is a core set of activities to regularly plot and review the price data as they are made
available by a market information system. These methods are the same as those presented
in the previous chapter to review the historical price data. The orientation in this section is
to provide guidance as to the actual steps that need to be done when monitoring prices.
In an earlier part of this manual there were suggestions on determining the commodities and
markets that should be monitored. Whether the monitoring is done intensively or
extensively, the same procedures should be routinely followed to analyze price data. The
Plot the price data by market commodity for the current year and a reference yea
(stacked time series graph).
This will give an indication season price pattern and
whether the price behavior is following or deviating from normal behavior.
Plot the historical time series price data in two ways.
First, graph specific
commodities for individual markets. Second, graph individual commodities across
many markets. These provide the historical basis for comparison and putting the
current data into context.
Plot the price data by market for all of the commodities being monitored in that
This will give an indication of how prices within a market are moving
For both sets of plots done in steps 2 and 3, answer the following question: Are
there any patterns that appear to be indicating a deviation from the expected pattern?
Finally, the rate of price change from one month to the next should be calculated and
compared to reference periods (e.g., last year and an average year). The rate of
change should be evaluated to see if it is about the same, faster, or slower than the