A demand analysis has been conducted for the demand of broiler chicken production in USA. The increasing broiler chicken demand in US certainly causing a deep thinking for the producers that which way to go i.e. either increases the production to meet the demand or increase the prices to control the demand by lowering it to an extent. For this purpose, several variables have been used to analyze the demand model for the broiler chicken production. Following are the variables used in our analysis which shall enable us to determine the factors that affects the broiler chicken demand in US:
BCP = Broiler Chicken Production (1000 Pounds)
BCPR = Broiler Chicken Price (Cents/Pound)
TCP = Turkey Chicken Production (1000 Pounds)
TCPR = Turkey Chicken Price (Cents/Pound)
RGI = Real Gross Income (1000 Dollars)
Transforming the variables in demand equation and re-estimating the Demand Function
Univariate demand analysis has been explained in the previous section where every variable explains its impact on the broiler chicken production independently. After obtaining highly significant results from the univariate analysis, multivariate regression analysis will be performed with combining all the independent variables under one model and then analyzing its importance and significance independently. The demand regression model used in the analysis is log - linear model i.e. of the following form:
Analysis of the transformed log linear model
The result of the correlation analysis (see table 1) of the overall linear regression model shows extremely significant linear relationship with the independent variables i.e. the independent variables are explaining almost 100% variation in the changes in Broiler chicken production. The obtained results from the regression analysis are certainly rarely found because it is hardly been observed that variation in independent variables explains a complete variation in the dependent variable. The adjusted R - square also plays an important role in the analysis of several variables when used as the independent variables. It imposes the penalty for the increment of every irrelevant variable and the resulting adjusted R - square for such a model which includes irrelevant variables may too low but here there is no difference between the R - square and the adjusted R - square i.e. it is also equal to 1.
Table 1: Model Summary
Model
R
R Square
Adjusted R Square
Std. Error of the Estimate
1
1.000a
1.000
1.000
.01293
a. Predictors: (Constant), GROSS INCOME (1000 DOLLARS), TURKEY FEED-PRICE RATIO, TURKEYCHICKEN EXPORTS (1000 POUNDS), TURKEY CHICKEN PRICE (CENTS/POUND), BROILER FEED-PRICE RATIO, BROILER CHICKEN PRICE (CENTS/POUND), YOUNG CHICKEN EXPORTS (1000 POUNDS), YOUNG CHICKEN CONSUMPTION (1000 POUNDS), TURKEY CHICKEN PRODUCTION (1000 POUNDS), CONSUMER PRICE INDEX (ALL POULTRY), TURKEY CHIKEN CONSUMPTION (1000 POUNDS)
b. Dependent Variable: Broiler chicken production (1000 POUNDS)
The use of ANOVA in regression analysis is to determine the significant variation in the overall model. The results obtained from table 2 shows that the overall model is highly significant as [f = 9.039E3, p = 0.000] which shows that variation in each variable causes a significant in the Broiler chicken production in USA.
Table 2: ANOVA
Model
Sum of Squares
df
Mean Square
F
Sig.
1
Regression
16.618
11
1.511
9.039E3
.000a
Residual
.006
33
.000
Total
16.623
44
a. Predictors: (Constant), GROSS INCOME (1000 DOLLARS), TURKEY FEED-PRICE ...