PREDICTING EGG PRODUCTION IN WHITE LEGHORNS 279 



nd March the error lies between one and two percent. When April, 

 lay, June, July, September and October records are used as a basis the 

 verage errors of prediction are about 2.50 to 4.50 percent of the average 

 nnual yield. 



The average deviations without regard to sign are less than 20 percent 

 f the annual production. The values for the individual months range 

 rom 15.7 for September to 19.1 for January. 



The square root of mean square deviations are less than 25 percent of 

 he average annual production. The individual values range from 20.9 

 Dr September to 24.6 for January. 



These two latter tests may at first seem to indicate very unsatisfactory 

 rediction. Such, however, is not the case. These give the average 

 rrors either above or below the true record made in the prediction of the results 

 jr an individual bird. The thing which is required in practice is generally 

 ie prediction for a group of birds of a particular class. In a flock of 415 

 irds this has been shown above to be possible with an error of less than 5 

 ercent of the actual production for any month of the year and less than one 

 er cent for a number of the months. 



The closeness of prediction may be made clear by a set of diagrams. 



In these the estimated production is shown by the straight line. The 

 ctual average production for the year or for the group of remaining months 

 )r which prediction is made is shown by solid dots for each group of birds 

 s classified by monthly record. The shaded areas are determined as 

 )llows. The birds were first grouped into classes of five-eggs range with 

 aspect to number of eggs laid during the period of time used as a basis 

 f prediction. The birds of these classes of five-eggs range were further 

 ibdivided into those in which actual egg production was greater than the 

 redicted and those in which the actual number was less than the predicted 

 umber. 4 The average error of prediction was determined for each of 

 icse groups, and these averages represent the upper and the lower limits 

 f the shaded areas. The upper limit represents, therefore, the average 

 eviation (for the period for which prediction is made) of all birds which 

 lake a higher record than that predicted for their class. The lower limit 



4 A range of five eggs was used in order to obtain a number of birds sufficiently large to reduce 

 inewhat the irregularities due to the errors of random sampling. The errors of prediction were 



each case determined for classes of unit range. Grouping is used for graphic representation 

 erely. The average deviations represented by the limits of the shaded zone are to be thought 



as measured from a line perpendicular to the ordinates and intersecting the prediction line 

 i the mid-ordinate of the 5-egg class. 



;NEDCS 6: My 1921 



