# Population Growth and Youth Unemployment in Kenya

Pages: 5 (1165 words) Published: December 10, 2012
Analytical models for hypothesis testing
Hypothesis H1 and H2
(i) For example, correlation modeling will apply for the external environment with the following independent variables: customers, competitors, suppliers, substitute products and demographic characteristics. The dependent variable will be the accuracy of performance forecasting. The expressions of the variables will be as indicated below: Dependent variable: Accuracy of performance forecast is denoted as A. Independent variables: Customers, denoted as Yi;

Competitors, denoted as Yj;
Suppliers, denoted as Yk;
Substitute products, denoted as Yl;
Demographic characteristics, denoted as Ym.
α – Constant term;
β – Beta coefficient;
ε – Error term.
The general regression models for each of the variables above will appear as follows as they get combined with one another: (1) Ai = α + βiYi + ε;
(2) Aj = α + βiYi + βjYj + ε;
(3) Ak = α + βiYi + βjYj + βkYk + ε;
(4) Al = α + βiYi + βjYj + βkYk + βlYl + ε
(5) Am = α + βiYi + βjYj + βkYk + βlYl + βmYm +ε

(ii) Causal models such as Y = a0 + a1X1 + . . + an Xn , where Y is the quantity to be forecasted and (X1, X2, . ., Xn) are n variables that have predictive power for Y.

Hypothesis H3 and H4
(i) For example for H4, The expressions of the variables will be as indicated below: Dependent variable: Accuracy of performance forecast is denoted as “A”. Independent variables: Strategy, denoted as Xi;

Moderating variables: Leadership, denoted as Xj;
Culture, denoted as Xk;
Structure, denoted as Xl;
α – Constants term
β – Beta coefficients;
ε – Error term.
The general regression models for each of the variables above will appear as follows as they get combined with one another: (1) Ai = α + βiXi + ε;
(2) Aj = α + βiXi + βjXj + ε;
(3) Ak = α + βiXi + βjXj + βkXk + ε;
(4) Al = α + βiXi + βjXj + βkXk + βlXl + ε

(ii) Causal models such as Y = a0 + a1X1 + . . + an Xn , where Y is the quantity to be forecasted and (X1, X2, . ., Xn) are n variables that have predictive power for Y.

Hypotheses H5 and H6
(i) For a response Y and two variables x1 and moderating variable x2, the regression model will be as follows:

In this case, the role of x2 as a moderating variable is accomplished by evaluating b3, the parameter estimate for the interaction term.

(ii) Regression model will apply for the moderating effect of the internal operating environment on the relationship between a forecasting method and accuracy. The expressions of the variables will be as indicated below: Dependent variable: Forecasting method, denoted as M.

Independent variables: Customers, denoted as Y1;
Competitors, denoted as Y2;
Suppliers, denoted as Y3;
Substitute products, denoted as Y4;
Demographic characteristics, denoted as Y5.
α – Constant term;
β – Beta coefficient;
ε – Error term.
The general regression models for each of the variables above will appear as follows as they get combined with one another: (1) M1 = α + β1Y1 + ε;
(2) M2 = α + β1Y1 + β2Y2 + ε;
(3) M3 = α + β1Y1 + β2Y2 + β3Y3 + ε;
(4) M4 = α + β1Y1 + β2Y2 + β3Y3 + β4Y4 + ε
(5) M5 = α + β1Y1 + β2Y2 + β3Y3 + β4Y4 + β5Y5 +ε

Hypothesis H7
Combined forecasting models, the independent variables xi ……….xn, will be as follows: x1 = Cost of forecasting method compared with its gains
x2 = Complexity of the relationships among variables
x3 = Time period involved
x4 = Accuracy needed to forecast
x5 = Lead time between receiving information and decision to be made x6 = By how we can catch changes and anomalies in data
x7 = By how easily the forecast can be altered as economy changes Regression equations will assume the pattern of H2 or H2 above.

In combining forecasting techniques in hypothesis H7, factor analysis will be used in determining parameters that are considered important in combining forecasts. The degree of importance is determined by the degree of factor loading of each parameter. The parameters...