38 BULLETIN 1440, L\ S. DEPARTMENT OF AGRICULTURE 
quantities which would have applied to only the central year of the 
period. 
The trend of demand is obtained from equation 2 by the following 
process: 
Holding all the factors except supply, price, population, and 
time constant, the equation (in terms of deviations from the means) 
becomes : 
Log x u = -0.62061 log xj + 2.60325 log x 6 + 0.02428 x 9 
and for a price constant at the average price, 
0= -0.62061 log Xi + 2.60325 log x 6 -f 0.02428 x 9 
rearranging 
(5) 0.62061 log x 1 =2.60325 log x 6 +0.02428 x 9 . 
Equation 5 gives the supply which, at any given time, can be sold 
for the average price, assuming that business activity, the price of 
steers, and European demand are all average. Putting the variables 
back into terms of whole numbers instead of deviations from the 
means 18 the equation becomes: 
r „ x ^ 2.60325 log X c + 0.02428 X 9 -208.60798 
° g " 1 0.62061 
Substituting the values of X 6 and X 9 for the different years in this 
equation, the upward trend used in plotting Figure 25 is obtained. 19 
Besides the two correlations mentioned above, several other sets 
were solved. In one, hog prices divided by the price index was sub- 
stituted for the raw prices, and these deflated prices (X 12 ) correlated 
with the whole set of 10 independent variables first used, leaving 
out only the price level (Xi" ). This gave a multiple correlation of 
B-12. 1,2,3,4,5,6,7,8,9=0.895, for the period covering 1907 to June, 1914. 
Correlation of the same variables for the postwar period from 
December, 1921, through June, 1924, gave a multiple correlation of 
K = 0.88. In this case, however,. the demand seemed to have a de- 
cidedly downward trend, both time and population showing negative 
regressions. 
Combining the pre-war and post-war periods (1907 through June, 
1914, and 1921 through June, 1924), and computing the multiple 
correlation for the seven independent variables used in obtaining 
the pre-war demand curve, a coefficient of 11 = 0.861 was obtained, 
as compared with R = 0.862 for the same factors for the pre-war 
period alone. The demand curve was practically identical with that 
obtained from the pre-war solution but the trend of demand was 
somewhat different, showing an upward trend for the pre-war years 
and a downward trend for the post-war years, as the net result of a 
positive net regression on population and a negative net regression 
on time. The various studies of the post-war demand lead to the 
conclusion that the steady upward trend in demand which charac- 
terized the pre-war period had (in 1924) not yet reappeared after 
the abrupt break due to the abnormal war conditions, and that the 
level of demand has been fluctuating since the war. Not until a 
longer period of post-war observations is available will it be possible 
to determine the present trend of the demand for hogs with the pre- 
cision with which the pre-war trend can be stated. 
1 5 For the method of computing the constant in this process, see (23) r>. 11, vote 7. 
19 Since the correlation from which these regression coefficients were obtained was not very near perfect 
(R=0.862), the change in price with changes in supply when reversed as just indicated, does not give a 
very accurate description of changes in quantity demanded with changes in prices asked. See (33, p. 526- 
638). 
