(Difference between revisions)
| Revision as of 15:34, 18 August 2006 Peter (Talk | contribs) (→'''Analyzing trends''') ← Previous diff |
Revision as of 19:35, 19 August 2006 Peter (Talk | contribs) (→'''Analyzing trends''') Next diff → |
||
| Line 1: | Line 1: | ||
| <blockquote> | <blockquote> | ||
| - | ='''Analyzing trends'''= | + | ==Analyzing trends== |
| The method uses historical data on crop area and production for up to 30 years to fit a linear regression model with area or production of the crop as dependent variable and time as the independent variable. | The method uses historical data on crop area and production for up to 30 years to fit a linear regression model with area or production of the crop as dependent variable and time as the independent variable. | ||
Revision as of 19:35, 19 August 2006
Analyzing trends
The method uses historical data on crop area and production for up to 30 years to fit a linear regression model with area or production of the crop as dependent variable and time as the independent variable. What this means basically is that the average yearly increase which was found over previous years is assumed to continue for future years.
An increase in crop yield from year to year may be explained by gradual improvements in technology, while other variables such as weather conditions, pests and diseases, input availability are all fairly constant or “under control”. A gradual increase in crop area may be explained by new land clearing or a gradual replacement of one crop by an other (for example sorghum/ maize in East Africa). This very simple model does not take into consideration any specific events of the current year, either physical or economic factors.