You are on page 1of 11

1

2
FORECASTING
Forecasting refers to the estimation of the value of the dependent variable , given the actual
or projected value of the independent variable . The forecast-error variance of is given by:

where is the number of observations and is the variance of . Since is seldom known, we
use as an unbiased estimate of , so that the estimated forecast-error variance, , is

3
The 95% confidence interval for the forecast , is

where and refers to the distribution with degrees of freedom.

4
5
6
7
MATLAB IMPLEMENTATION OF THE PROBLEM
Output:
(a)
Input:
X=15;
YF=59.13-2.60*X; CI=(53.8568, -13.5968)
n=15;
X_bar=11.00;
sum_x2=442;
s2=220.99;
s2F=s2*(1+1/n+(X-X_bar)^2/sum_x2);
sF=sqrt(s2F);
t025=tinv(1-0.025,13);
Yu=YF+t025*sF;
Yl=YF-t025*sF;
fprintf('YF=%g\n',YF)
fprintf('t025=%g\n',t025)
fprintf('CI=(%g, %g)\n',Yu,Yl)
8
(b) Output:
Input:
X=11.5;
YF=59.13-2.60*X;
n=15;
X_bar=11.00;
sum_x2=442;
s2=220.99;
s2F=s2*(1+1/n+(X-X_bar)^2/sum_x2);
sF=sqrt(s2F);
t025=tinv(1-0.025,13);
Yu=YF+t025*sF;
Yl=YF-t025*sF;
fprintf('YF=%g\n',YF)
fprintf('t025=%g\n',t025)
fprintf('CI=(%g, %g)\n',Yu,Yl)

9
Find the 95% confidence interval of for and in 1981, given that , , .
,,,

and

The 95% confidence interval for , in 1981 is then or between and .

10
BEST OF LUCK

11

You might also like