User Tools

Site Tools

capri:concept:referencerun

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
capri:concept:referencerun [2011/04/13 16:16] – gochtcapri:concept:referencerun [2022/11/07 10:23] (current) – external edit 127.0.0.1
Line 1: Line 1:
-<<!-- Start Inhalt --> +p
-    <div class=rightwindow> +
-       +
-      <H2>Reference run</H2> +
-      <H3>Overview</H3> +
-      <TABLE padding="20px"><TR><TD> +
-      <P>The reference run or baseline serves as a reference point for ex-ante impact analysis +
- +
-      with CAPRI. It reflects the most probable development in agricultural markets +
- +
-      from global to regional scale for 8-10 year time horizon, at the current legislation. +
- +
-      It integrates agricultural market projections from other insitutions as FAPRI, FAO and DG-AGRI. +
- +
-      Unique for CAPRI are regional resolution below the national level for EU27 at the level of +
- +
-      NUTS 2 regions, and the Bayesian methodology applied.</P> +
- +
- +
- +
-      <P>The <A HREF="supply.htm">supply</A> and <A HREF="market.htm">market</A> module +
- +
-      are calibrated to results of the reference run.</P> +
- +
- +
- +
- +
- +
- +
- +
- +
- +
- +
- +
-      <P>In opposite to many other reference run approaches, CAPRI employs a Bayesian estimation +
- +
-      framework to define a mutually consistent set of projection values for activity levels, +
- +
-      yields, production, feed and processing demand, human consumption, trade etc. Basically, +
- +
-      all major equations from the supply and market modules are defined as in an optimization +
- +
-      framework which maximizes the joint posterior density for given a priori distribution +
- +
-      of the different elements.</P> +
- +
- +
- +
-      </TD> +
- +
-      <TD><P><a href="images\refrun.gif"><img style="text-decoration: none;border: none;color: white" src="images\refrun.gif" width="500px"/></a></P> +
- +
-      </TD> +
- +
-      </TR> +
- +
-      </TABLE> +
- +
- +
- +
-      <H3>Methodology</H3> +
- +
- +
- +
-      <TABLE padding="10 px"><TR> +
- +
-      <TD><P><a href="images\support.jpg"><img style="text-decoration: none;border: none;color: white;padding:20px;margin:20px;" src="images\support.jpg" width="500px"/></a></P> +
- +
-      </TD> +
- +
-      <TD> +
- +
- +
- +
-      <P>The a priori distributions are derived from two different sources. For all elements, +
- +
-      long terms are projected. Using "no change" as the zero hypothesis, a weighted average +
- +
-      between the base year value and the trend estimate is calculated, using R squared +
- +
-      as weight for the trend estimate.</P> +
- +
-      <P>The resulting independent projections are then corrected +
- +
-      by the relative effect of implementing already decided upon changes in agricultural policies in the base +
- +
-      year - so-called "policy shifts". These effects are derived from a simulation with the full modelling system. The +
- +
-      resulting estimates along with the estimation error of the trend deliver a priori distributions +
- +
-      for all items in the estimation framework.</P> +
- +
- +
- +
- +
- +
- +
- +
-      <P>In a first step, the estimation is solved for these trend support, indepedently for each country. Afterwards, results are aggregated to +
- +
-      EU. The projections of other instutions are then added, and replace where available the supports based on trends. +
- +
-      If the projections are only reporting values for the EU and not for individual countries, +
- +
-      the country results from the first steps are used to distributed the EU estimate. The standard errors for the there projections, +
- +
-      especially for those provided by DG-AGRI, is set rather narrow, ensuring that the values are recovered as long as they do not violate +
- +
-      the consistency restrictions.</P> +
- +
- +
- +
-     <P>The estimation is then repeated in a second step based on the updated supports. The country results are then taken is given, and broken +
- +
-     down in a similar framework to NUTS 2 level, and from there to farm type groups.</P> +
- +
- +
- +
-      </TD> +
- +
-      </TR> +
- +
-      </TABLE> +
- +
- +
- +
-     <H3>Quality control</H3> +
- +
-     <P>Due to the NUTS 2 resolution with its 250 regions, and the high number of activities, outputs and inputs covered by the CAPRI modelling +
- +
-     system, the references comprises several Mio numbers. Therefore, aggregates over activities (single cereals aggregated to all cereals) +
- +
-     and products, and over countries to EU15, EU10 etc. are defined and reported in tables accessible via the Graphical user interface. +
- +
-     Those tables compare systematically the different inputs (trend estimate, policy shift, estimation results from the different steps).</P> +
- +
- +
- +
-     <P>Nevertheless, checking results for plausibility remains a challenge. An interested comparison of the methdology between the reference +
- +
-     run exercise for the OECD/FAO AGLINK-COSIMA model with an estimated input of 20 person months and the CAPRI reference run +
- +
-     with an estimated inputs of about 2 person months is reported by <A HREF=http://purl.umn.edu/44120>Adenauer 2008</A>.</P> +
- +
- +
- +
-<H3>More information</H3> +
- +
-<P class="pubparagraph"><span class="pubAuthor">Witzke H.P. and Adenaeuer M.</span>:<BR> +
- +
-<I><a HREF="docs\capri_baseline.pdf" class="intext">Capri Baseline (pdf, 13 slides)</a></I></p> +
- +
-</div><!-- Ende RightWindow --> +
- +
-      </div><!-- Ende content --> +
- +
- +
-</HTML> +
capri/concept/referencerun.1302711405.txt · Last modified: 2022/11/07 10:23 (external edit)