y1 |
y2 |
|
x1 |
x2 |
|
|
|
|
-2 |
1 |
|
0 |
-1 |
|
|
|
|
0 |
0 |
|
0 |
0 |
|
|
|
|
0 |
-1 |
|
0 |
0 |
|
|
|
|
1 |
3 |
|
0 |
1 |
|
|
|
|
0 |
-1 |
|
1 |
0 |
|
|
|
|
1 |
1 |
|
-1 |
0 |
|
|
|
|
|
|
|
|
|
|
|
|
|
y1 = b21 y2 + g11 x1 + u1 |
|
|
|
|
|
|
|
|
y2 = b12 y1 + g22 x2 + u2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-2 |
1 |
|
|
0 |
-1 |
|
|
|
0 |
0 |
|
|
0 |
0 |
|
|
Y= |
0 |
-1 |
|
X= |
0 |
0 |
|
|
|
1 |
3 |
|
|
0 |
1 |
|
|
|
0 |
-1 |
|
|
1 |
0 |
|
|
|
1 |
1 |
|
|
-1 |
0 |
|
|
|
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
P^=(XTX)-1XTY= |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
XTX= |
2 |
0 |
|
(XTX)-1= |
0,5 |
0 |
|
|
|
0 |
2 |
|
|
0 |
0,5 |
|
|
Wyznacznik (XTX)= |
|
4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
XTY= |
-1 |
-2 |
|
|
|
|
|
|
|
3 |
2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
P^=(XTX)-1XTY= |
|
-0,5 |
-1 |
|
|
|
|
|
|
|
1,5 |
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pośrednia MNK do I równania |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
b21= |
1,5 |
|
b12= |
2 |
|
|
|
|
g11= |
1 |
|
g22= |
-2 |
|
|
|
|
Takie same oceny otrzymamy stosując 2MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2MNK do I równania |
|
|
|
|
|
|
|
|
I Etap |
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji. Otrzymujemy wartości teoretyczne Y. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Y^=XTP^= |
-1,5 |
-1 |
|
|
|
|
|
|
|
0 |
0 |
|
|
|
|
|
|
|
0 |
0 |
|
|
|
|
|
|
|
1,5 |
1 |
|
|
|
|
|
|
|
-0,5 |
-1 |
|
|
|
|
|
|
|
0,5 |
1 |
|
|
|
|
|
|
|
macierz wartości teoretycznych |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II Etap |
|
|
|
|
|
|
|
|
Do I równania po wstawieniu wartości teoretycznych za y2 stosujemy zwykłą MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
y1 = b21 y2^ + g11 x1 + v1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-1 |
0 |
|
|
-2 |
|
|
|
Z= |
0 |
0 |
|
|
0 |
|
4 |
|
|
0 |
0 |
|
y= |
0 |
ZTy= |
-1 |
|
|
1 |
0 |
|
|
1 |
|
|
|
|
-1 |
1 |
|
|
0 |
|
|
|
|
1 |
-1 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
(ZTZ)= |
4 |
-2 |
|
|
|
(ZTZ)-1= |
0,5 |
0,5 |
|
-2 |
2 |
|
|
|
|
0,5 |
1 |
|
wyznacz= |
4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
1,5 |
oceny parametrów pierwszego równania |
|
|
|
|
|
|
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chcemy policzyć błędy średnie szacunku |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
w. Reszt |
|
|
|
u^2= |
|
|
|
|
u^= |
-3,5 |
|
|
12,25 |
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
1,5 |
|
|
2,25 |
|
|
|
|
|
-3,5 |
|
|
12,25 |
|
|
|
|
|
0,5 |
|
|
0,25 |
|
|
|
|
|
0,5 |
|
|
0,25 |
|
|
|
|
|
|
|
|
27,25 |
suma kw reszt |
|
|
|
Wariancja resztowa |
|
s2= |
6,8125 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ocena asymptotycznej macierzy kowariancji estymatora 2MNK: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
s2(ZTZ)-1= |
3,40625 |
3,40625 |
|
|
|
|
|
|
|
3,40625 |
6,8125 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Błędy średnie szacunku: |
|
|
|
|
|
|
|
|
b21 |
1,84560288252918 |
|
|
|
|
|
|
|
g11 |
2,61007662722764 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II równanie |
|
|
|
|
|
|
|
|
Jest jednoznacznie identyfikowalne Pośrednia MNK = 2MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2MNK do II równania |
|
|
|
|
|
|
|
|
I Etap |
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji. Otrzymujemy wartości teoretyczne Y. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II Etap |
|
|
|
|
|
|
|
|
Do II równania po wstawieniu wartości teoretycznych za y1 stosujemy zwykłą MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
y2 = b12 y1^ + g22 x2 + v2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-1,5 |
-1 |
|
|
1 |
|
|
|
Z= |
0 |
0 |
|
|
0 |
|
4 |
|
|
0 |
0 |
|
y= |
-1 |
ZTy= |
2 |
|
|
1,5 |
1 |
|
|
3 |
|
|
|
|
-0,5 |
0 |
|
|
-1 |
|
|
|
|
0,5 |
0 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
(ZTZ)= |
5 |
3 |
|
|
|
(ZTZ)-1= |
2 |
-3 |
|
3 |
2 |
|
|
|
|
-3 |
5 |
|
wyznacz= |
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
2 |
oceny parametrów drugiego równania |
|
|
|
|
|
|
|
-2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chcemy policzyć błędy średnie szacunku |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
w. Reszt |
|
|
|
u^2= |
|
|
|
|
u^= |
3 |
|
|
8,99999999999999 |
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
-1 |
|
|
1 |
|
|
|
|
|
3 |
|
|
9 |
|
|
|
|
|
-1 |
|
|
1 |
|
|
|
|
|
-0,999999999999998 |
|
|
0,999999999999996 |
|
|
|
|
|
|
|
|
21 |
suma kw reszt |
|
|
|
Wariancja resztowa |
|
s2= |
5,25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ocena asymptotycznej macierzy kowariancji estymatora 2MNK: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
s2(ZTZ)-1= |
10,5 |
-15,75 |
|
|
|
|
|
|
|
-15,75 |
26,25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Błędy średnie szacunku: |
|
|
|
|
|
|
|
|
b21 |
3,24037034920393 |
|
|
|
|
|
|
|
g11 |
5,1234753829798 |
|
|
|
|
|
|
|
y1 |
y2 |
|
x1 |
x2 |
UWAGA: |
Zakładamy, że mamy |
|
|
-2 |
1 |
|
0 |
-1 |
|
do czynienia z |
|
|
0 |
0 |
|
0 |
0 |
|
modelem rekurencyjnym |
|
|
0 |
-1 |
|
0 |
0 |
|
|
|
|
1 |
3 |
|
0 |
1 |
|
|
|
|
0 |
-1 |
|
1 |
0 |
|
|
|
|
1 |
1 |
|
-1 |
0 |
|
|
|
|
|
|
|
|
|
|
|
|
|
y1 = g11 x1 + g31 + u1 |
|
|
|
|
|
|
|
|
y2 = b12 y1 + g22 x2 + u2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
To nie jest potrzebne |
|
|
|
|
|
|
|
|
|
-2 |
1 |
|
|
0 |
-1 |
1 |
|
|
0 |
0 |
|
|
0 |
0 |
1 |
|
Y= |
0 |
-1 |
|
X= |
0 |
0 |
1 |
|
|
1 |
3 |
|
|
0 |
1 |
1 |
|
|
0 |
-1 |
|
|
1 |
0 |
1 |
|
|
1 |
1 |
|
|
-1 |
0 |
1 |
|
|
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
P^=(XTX)-1XTY= |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
0 |
0 |
|
0,5 |
0 |
0 |
|
XTX= |
0 |
2 |
0 |
(XTX)-1= |
0 |
0,5 |
0 |
|
|
0 |
0 |
6 |
|
0 |
0 |
0,166666666666667 |
|
Wyznacznik (XTX)= |
|
24 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
XTY= |
-1 |
-2 |
|
|
|
|
|
|
|
3 |
2 |
|
|
|
|
|
|
|
0 |
3 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
P^=(XTX)-1XTY= |
|
-0,5 |
-1 |
|
|
|
|
|
|
|
1,5 |
1 |
|
|
|
|
|
|
|
0 |
0,5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pośrednia MNK do I równania |
|
|
równania są niejedn. Identyf. Nie można stosować Pośredniej MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
g31= |
0 |
|
b12= |
2 |
|
|
|
|
g11= |
-0,5 |
|
g22= |
-2 |
|
|
|
|
Takie same oceny otrzymamy stosując 2MNK - tu to nie jest prawdą |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2MNK do I równania |
|
to samo co MNK, bo równ. Oderwane |
|
|
|
|
|
|
I Etap |
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji. Otrzymujemy wartości teoretyczne Y. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Y^=XTP^= |
-1,5 |
-0,5 |
|
|
|
|
|
|
|
0 |
0,5 |
|
|
|
|
|
|
|
0 |
0,5 |
|
|
|
|
|
|
|
1,5 |
1,5 |
|
|
|
|
|
|
|
-0,5 |
-0,5 |
|
|
|
|
|
|
|
0,5 |
1,5 |
|
|
|
|
|
|
|
macierz wartości teoretycznych |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II Etap |
|
|
|
|
|
|
|
|
Do I równania stosujemy zwykłą MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
y1 = g31 + g11 x1 + v1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
0 |
|
|
-2 |
|
|
|
Z= |
1 |
0 |
|
|
0 |
|
0 |
|
|
1 |
0 |
|
y= |
0 |
ZTy= |
-1 |
|
|
1 |
0 |
|
|
1 |
|
|
|
|
1 |
1 |
|
|
0 |
|
|
|
|
1 |
-1 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
(ZTZ)= |
6 |
0 |
|
|
|
(ZTZ)-1= |
0,166666666666667 |
0 |
|
0 |
2 |
|
|
|
|
0 |
0,5 |
|
wyznacz= |
12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
0 |
oceny parametrów pierwszego równania |
|
|
|
|
|
|
|
-0,5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chcemy policzyć błędy średnie szacunku |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
w. Reszt |
|
|
|
u^2= |
|
|
|
|
u^= |
-2 |
|
|
4 |
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
1 |
|
|
1 |
|
|
|
|
|
0,5 |
|
|
0,25 |
|
|
|
|
|
0,5 |
|
|
0,25 |
|
|
|
|
|
|
|
|
5,5 |
suma kw reszt |
|
|
|
Wariancja resztowa |
|
s2= |
1,375 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ocena asymptotycznej macierzy kowariancji estymatora 2MNK: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
s2(ZTZ)-1= |
0,229166666666667 |
0 |
|
|
|
|
|
|
|
0 |
0,6875 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Błędy średnie szacunku: |
|
|
|
|
|
|
|
|
g31 |
0,478713553878169 |
|
|
|
|
|
|
|
g11 |
0,82915619758885 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II równanie |
|
|
|
|
|
|
|
|
stasujemy MNK |
|
jako rek |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
y2 = b12 y1 + g22 x2 + v2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-2 |
-1 |
|
|
1 |
|
|
|
Z= |
0 |
0 |
|
|
0 |
|
2 |
|
|
0 |
0 |
|
y= |
-1 |
ZTy= |
2 |
|
|
1 |
1 |
|
|
3 |
|
|
|
|
0 |
0 |
|
|
-1 |
|
|
|
|
1 |
0 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
(ZTZ)= |
6 |
3 |
|
|
|
(ZTZ)-1= |
0,666666666666667 |
-1 |
|
3 |
2 |
|
|
|
|
-1 |
2 |
|
wyznacz= |
3 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
-0,666666666666667 |
oceny parametrów drugiego równania |
|
|
|
|
|
|
|
2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chcemy policzyć błędy średnie szacunku |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
w. Reszt |
|
|
|
u^2= |
|
|
|
|
u^= |
1,66666666666667 |
|
|
2,77777777777778 |
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
-1 |
|
|
1 |
|
|
|
|
|
1,66666666666667 |
|
|
2,77777777777778 |
|
|
|
|
|
-1 |
|
|
1 |
|
|
|
|
|
1,66666666666667 |
|
|
2,77777777777778 |
|
|
|
|
|
|
|
|
10,3333333333333 |
suma kw reszt |
|
|
|
Wariancja resztowa |
|
s2= |
2,58333333333333 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ocena asymptotycznej macierzy kowariancji estymatora 2MNK: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
s2(ZTZ)-1= |
1,72222222222222 |
-2,58333333333333 |
|
|
|
|
|
|
|
-2,58333333333333 |
5,16666666666667 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Błędy średnie szacunku: |
|
|
|
|
|
|
|
|
b12 |
1,31233464566864 |
|
|
|
|
|
|
|
g22 |
2,27303028283098 |
|
|
|
|
|
|
|
y1 |
y2 |
|
x1 |
x2 |
|
2mnk |
jeśli model o rownaniach wspolzaleznych |
|
-2 |
1 |
|
0 |
-1 |
|
|
|
|
0 |
0 |
|
0 |
0 |
|
|
|
|
0 |
-1 |
|
0 |
0 |
|
|
|
|
1 |
3 |
|
0 |
1 |
|
|
|
|
0 |
-1 |
|
1 |
0 |
|
|
|
|
1 |
1 |
|
-1 |
0 |
|
|
|
|
|
|
|
|
|
|
|
|
|
y1 = g11 x1 + g31 + u1 |
|
|
|
|
|
|
|
|
y2 = b12 y1 + g22 x2 + u2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-2 |
1 |
|
|
0 |
-1 |
1 |
|
|
0 |
0 |
|
|
0 |
0 |
1 |
|
Y= |
0 |
-1 |
|
X= |
0 |
0 |
1 |
|
|
1 |
3 |
|
|
0 |
1 |
1 |
|
|
0 |
-1 |
|
|
1 |
0 |
1 |
|
|
1 |
1 |
|
|
-1 |
0 |
1 |
|
|
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
P^=(XTX)-1XTY= |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
0 |
0 |
|
0,5 |
0 |
0 |
|
XTX= |
0 |
2 |
0 |
(XTX)-1= |
0 |
0,5 |
0 |
|
|
0 |
0 |
6 |
|
0 |
0 |
0,166666666666667 |
|
Wyznacznik (XTX)= |
|
24 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
XTY= |
-1 |
-2 |
|
|
|
|
|
|
|
3 |
2 |
|
|
|
|
|
|
|
0 |
3 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
P^=(XTX)-1XTY= |
|
-0,5 |
-1 |
|
|
|
|
|
|
|
1,5 |
1 |
|
|
|
|
|
|
|
0 |
0,5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pośrednia MNK do I równania |
|
|
|
równ. Są niejednozn. Identyf - nie można stosować Poś. MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
g31= |
0 |
|
b12= |
2 |
|
|
|
|
g11= |
-0,5 |
|
g22= |
-2 |
|
|
|
|
Takie same oceny otrzymamy stosując 2MNK - tu to nie jest prawda |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2MNK do I równania |
|
|
|
|
|
|
|
|
I Etap |
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji. Otrzymujemy wartości teoretyczne Y. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Y^=XTP^= |
-1,5 |
-0,5 |
|
|
|
|
|
|
|
0 |
0,5 |
|
|
|
|
|
|
|
0 |
0,5 |
|
|
|
|
|
|
|
1,5 |
1,5 |
|
|
|
|
|
|
|
-0,5 |
-0,5 |
|
|
|
|
|
|
|
0,5 |
1,5 |
|
|
|
|
|
|
|
macierz wartości teoretycznych |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II Etap |
|
|
|
|
|
|
|
|
Do I równania 2MNK = MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
y1 = g31 + g11 x1 + v1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
0 |
|
|
-2 |
|
|
|
Z= |
1 |
0 |
|
|
0 |
|
0 |
|
|
1 |
0 |
|
y= |
0 |
ZTy= |
-1 |
|
|
1 |
0 |
|
|
1 |
|
|
|
|
1 |
1 |
|
|
0 |
|
|
|
|
1 |
-1 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
(ZTZ)= |
6 |
0 |
|
|
|
(ZTZ)-1= |
0,166666666666667 |
0 |
|
0 |
2 |
|
|
|
|
0 |
0,5 |
|
wyznacz= |
12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
0 |
oceny parametrów pierwszego równania |
|
|
|
|
|
|
|
-0,5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chcemy policzyć błędy średnie szacunku |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
w. Reszt |
|
|
|
u^2= |
|
|
|
|
u^= |
-2 |
|
|
4 |
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
1 |
|
|
1 |
|
|
|
|
|
0,5 |
|
|
0,25 |
|
|
|
|
|
0,5 |
|
|
0,25 |
|
|
|
|
|
|
|
|
5,5 |
suma kw reszt |
|
|
|
Wariancja resztowa |
|
s2= |
1,375 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ocena asymptotycznej macierzy kowariancji estymatora 2MNK: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
s2(ZTZ)-1= |
0,229166666666667 |
0 |
|
|
|
|
|
|
|
0 |
0,6875 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Błędy średnie szacunku: |
|
|
|
|
|
|
|
|
b21 |
0,478713553878169 |
|
|
|
|
|
|
|
g11 |
0,82915619758885 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II równanie |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2MNK do II równania |
|
|
|
|
|
|
|
|
I Etap |
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji. Otrzymujemy wartości teoretyczne Y. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II Etap |
|
|
|
|
|
|
|
|
Do II równania po wstawieniu wartości teoretycznych za y1 stosujemy zwykłą MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
y2 = b12 y1^ + g22 x2 + v2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
-1,5 |
-1 |
|
|
1 |
|
|
|
Z= |
0 |
0 |
|
|
0 |
|
4 |
|
|
0 |
0 |
|
y= |
-1 |
ZTy= |
2 |
|
|
1,5 |
1 |
|
|
3 |
|
|
|
|
-0,5 |
0 |
|
|
-1 |
|
|
|
|
0,5 |
0 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
|
|
(ZTZ)= |
5 |
3 |
|
|
|
(ZTZ)-1= |
2 |
-3 |
|
3 |
2 |
|
|
|
|
-3 |
5 |
|
wyznacz= |
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
2 |
oceny parametrów drugiego równania |
|
|
|
|
|
|
|
-2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chcemy policzyć błędy średnie szacunku |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
w. Reszt |
|
|
|
u^2= |
|
|
|
|
u^= |
3 |
|
|
8,99999999999999 |
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
-1 |
|
|
1 |
|
|
|
|
|
3 |
|
|
9 |
|
|
|
|
|
-1 |
|
|
1 |
|
|
|
|
|
-0,999999999999998 |
|
|
0,999999999999996 |
|
|
|
|
|
|
|
|
21 |
suma kw reszt |
|
|
|
Wariancja resztowa |
|
s2= |
5,25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ocena asymptotycznej macierzy kowariancji estymatora 2MNK: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
s2(ZTZ)-1= |
10,5 |
-15,75 |
|
|
|
|
|
|
|
-15,75 |
26,25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Błędy średnie szacunku: |
|
|
|
|
|
|
|
|
b21 |
3,24037034920393 |
|
|
|
|
|
|
|
g11 |
5,1234753829798 |
|
|
|
|
|
|
|
Modele wielorównaniowe - przykład |
|
|
|
|
|
|
|
|
|
|
(Goryl A., Z. Jędrzejczyk, K. Kukuła, J. Osiewalski "Wprowadzenie do ekonometrii w przykładach i zadaniach", PWN , Warszwa 1999) |
|
|
|
|
|
|
|
|
|
|
Dany jest model wielorównaniowy: |
|
|
|
|
|
|
|
|
|
|
yt1 = b21 yt2 + g11 xt1 + ut1 |
|
|
|
|
|
|
|
|
|
|
yt2 = b12 yt1 + b32 yt3 + ut2 |
|
|
|
|
|
|
|
|
|
|
yt3 = g23 xt2 + ut3 |
|
|
|
|
|
|
|
|
|
|
1. Przedstawić zapis macierzowy postaci strukturalnej modelu. |
|
|
|
|
|
|
|
|
|
|
2. Zbadać zupełność modelu oraz określić klasę modelu. |
|
|
|
|
|
|
|
|
|
|
3. Znaleźć postać zredukowaną z ograniczeniami. |
|
|
|
|
|
|
|
|
|
|
4. Zbadać identyfikowalność poszczególnych równań oraz określić właściwą metodę estymacji. |
|
|
|
|
|
|
|
|
|
|
5. Oszacować równania podając oceny parametrów i przybliżone błędy średnie szacunku. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Dane: |
|
|
|
|
|
|
|
|
|
|
yt1 |
yt2 |
yt3 |
xt1 |
xt2 |
|
|
|
|
|
|
2 |
2 |
3 |
1 |
0 |
|
|
|
|
|
|
0 |
0 |
1 |
0 |
0 |
|
|
|
|
|
|
1 |
0 |
-1 |
0 |
0 |
|
|
|
|
|
|
-3 |
-1 |
0 |
-1 |
0 |
|
|
|
|
|
|
1 |
0 |
-2 |
0 |
-1 |
|
|
|
|
|
|
-1 |
-1 |
-1 |
0 |
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1. Zapis macierzowy: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
yt = (yt1, yt2, yt3), |
|
xt = (xt1, xt2), |
|
ut = (ut1, ut2, ut3) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
yt B + xt G = ut |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
B= |
1 |
"-b12 " |
0 |
|
G= |
"-g11 " |
0 |
0 |
|
|
|
"-b21 " |
1 |
0 |
|
|
0 |
0 |
"-g23 " |
|
|
|
0 |
"-b32 " |
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
"-b12 " |
0 |
|
|
|
|
|
|
|
(yt1, yt2, yt3) |
"-b21 " |
1 |
0 |
+ |
(xt1, xt2) |
"-g11 " |
0 |
0 |
= |
ut |
|
0 |
"-b32 " |
1 |
|
|
0 |
0 |
"-g23 " |
|
|
|
|
|
|
|
|
|
|
|
|
|
2. Zupełność modelu: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
det(B)= |
1-b12*b21 |
|
|
|
|
|
|
|
|
|
Model jest zupełny, gdy b12*b21=/=1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Klasa modelu: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Model o równaniach współzależnych. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3. Postać zredukowana z restrykcjami: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
yt =xt P + vt |
|
P =-GB-1 |
|
vt = ut B-1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
P= |
p11 |
p12 |
p13 |
= |
g11/(1-b12b21) |
|
b12g11/(1-b12b21) |
|
0 |
|
|
p21 |
p22 |
p23 |
|
b32b21g23/(1-b12b21) |
|
b32g23/(1-b12b21) |
|
"g23 " |
|
|
|
|
|
|
|
|
|
|
|
|
Warunek poboczny: |
|
p13 =0 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4. Identyfikowalność równań: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Identyfikowalność I równania postaci strukturalnej: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pb(1) = -g(1) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
p11 |
p12 |
p13 |
* |
1 |
= |
"g11 " |
|
|
|
|
p21 |
p22 |
p23 |
|
"-b21 " |
|
0 |
|
|
|
|
|
|
|
|
0 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
b21 = |
p21 /p22 |
|
p22 =/=0 |
|
|
|
|
|
|
|
g11 = |
p11-p12* p21 /p22 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pierwsze równanie postaci strukturalnej jest jednoznacznie identyfikowalne. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Identyfikowalność II równania postaci strukturalnej: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pb(2) = -g(2) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
p11 |
p12 |
p13 |
* |
"-b12 " |
= |
0 |
|
|
|
|
p21 |
p22 |
p23 |
|
1 |
|
0 |
|
|
|
|
|
|
|
|
"-b32 " |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
b12 = |
(p12p23-p13p22)/(p11p23 - p13 p21 ) |
|
|
|
(p11p23 - p13 p21)=/=0 |
|
|
|
|
|
b32 = |
(p11p22-p12p21)/(p11p23 - p13 p21 ) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Drugie równanie postaci strukturalnej jest jednoznacznie identyfikowalne. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Identyfikowalność III równania postaci strukturalnej: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pb(3) = -g(3) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
p11 |
p12 |
p13 |
* |
0 |
= |
0 |
|
|
|
|
p21 |
p22 |
p23 |
|
0 |
|
"g23 " |
|
|
|
|
|
|
|
|
1 |
|
|
|
|
|
|
p23 = |
"g23 " |
|
|
|
|
|
|
|
|
|
p13 =0 |
fałszywe dla prawie wszystkich macierzy P |
|
|
|
|
|
|
|
|
|
Trzecie równanie postaci strukturalnej jest niejednoznacznie identyfikowalne. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5. Estymacja |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
2 |
3 |
|
1 |
0 |
|
|
|
|
|
0 |
0 |
1 |
|
0 |
0 |
|
|
|
|
Y= |
1 |
0 |
-1 |
X= |
0 |
0 |
|
|
|
|
|
-3 |
-1 |
0 |
|
-1 |
0 |
|
|
|
|
|
1 |
0 |
-2 |
|
0 |
-1 |
|
|
|
|
|
-1 |
-1 |
-1 |
|
0 |
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
P^=(XTX)-1XTY |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
XTX= |
2 |
0 |
|
(XTX)-1= |
0,5 |
0 |
|
|
|
|
|
0 |
2 |
|
|
0 |
0,5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
det(XTX)= |
4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
XTY= |
5 |
3 |
3 |
|
|
|
|
|
|
|
|
-2 |
-1 |
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
P^=(XTX)-1XTY= |
|
2,5 |
1,5 |
1,5 |
= |
p11 |
p12 |
p13 |
|
|
|
|
-1 |
-0,5 |
0,5 |
|
p21 |
p22 |
p23 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pośrednia MNK do I równania |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
b21 = |
p21 /p22 |
= |
2 |
|
|
|
|
|
|
|
g11 = |
p11-p12* p21 /p22 |
|
= |
-0,5 |
|
|
|
|
|
|
Takie same oceny otrzymamy stosując 2MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2MNK do I równania |
|
|
|
|
|
|
|
|
|
|
I Etap |
|
|
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji. Otrzymujemy wartości teoretyczne Y^: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
y^t1 |
y^t2 |
y^t3 |
|
|
|
|
|
|
|
Y^=XTP^= |
2,5 |
1,5 |
1,5 |
|
|
|
|
|
|
|
|
0 |
0 |
0 |
|
|
|
|
|
|
|
|
0 |
0 |
0 |
|
|
|
|
|
|
|
|
-2,5 |
-1,5 |
-1,5 |
|
|
|
|
|
|
|
|
1 |
0,5 |
-0,5 |
|
|
|
|
|
|
|
|
-1 |
-0,5 |
0,5 |
|
|
|
|
|
|
|
|
macierz wartości teoretycznych |
|
|
|
|
|
|
|
|
|
II Etap |
|
|
|
|
|
|
|
|
|
|
Do I równania po wstawieniu wartości teoretycznych za yt2 stosujemy zwykłą MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
yt1 = b21 y^t2 + g11 xt1 + wt1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(b21, g11)' |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
y^t2 |
xt1 |
|
|
yt1 |
|
|
|
|
|
|
1,5 |
1 |
|
|
2 |
|
|
|
|
|
Z= |
0 |
0 |
|
|
0 |
|
8,5 |
|
|
|
|
0 |
0 |
|
y= |
1 |
ZTy= |
5 |
|
|
|
|
-1,5 |
-1 |
|
|
-3 |
|
|
|
|
|
|
0,5 |
0 |
|
|
1 |
|
|
|
|
|
|
-0,5 |
0 |
|
|
-1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
(ZTZ)= |
5 |
3 |
|
(ZTZ)-1= |
2 |
-3 |
|
|
|
|
|
3 |
2 |
|
|
-3 |
5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
det(ZTZ)= |
1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
2 |
b21^ |
oceny parametrów pierwszego równania |
|
|
|
|
|
|
|
|
-0,499999999999996 |
g11^ |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chcemy policzyć błędy średnie szacunku |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
w. reszt |
u^t1=yt1 - b21^ yt2 - g11^ xt1 |
|
|
|
|
|
|
|
|
|
|
-1,5 |
|
|
2,24999999999999 |
|
|
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
|
u^= |
1 |
|
u^^2= |
1 |
|
|
|
|
|
|
|
-1,5 |
|
|
2,25 |
|
|
|
|
|
|
|
1 |
|
|
1 |
|
|
|
|
|
|
|
0,999999999999996 |
|
|
0,999999999999993 |
|
|
|
|
|
|
|
|
|
|
7,49999999999998 |
suma kw. reszt |
|
|
|
|
|
Wariancja resztowa |
|
s2= |
1,875 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ocena asymptotycznej macierzy kowariancji estymatora 2MNK: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
s2(ZTZ)-1= |
3,74999999999999 |
-5,62499999999999 |
|
|
|
|
|
|
|
|
|
-5,62499999999999 |
9,37499999999998 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Przybliżone błędy średnie szacunku: |
|
|
|
|
|
|
|
|
|
|
D(b21^)= |
1,93649167310371 |
|
|
|
|
|
|
|
|
|
D(g11^)= |
3,06186217847897 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II równanie |
|
|
|
|
|
|
|
|
|
|
Jest jednoznacznie identyfikowalne Pośrednia MNK = 2MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2MNK do II równania |
|
|
|
|
|
|
|
|
|
|
I Etap |
|
|
|
|
|
|
|
|
|
|
Szacujemy postać zredukowaną bez restrykcji. Otrzymujemy wartości teoretyczne Y. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
II Etap |
|
|
|
|
|
|
|
|
|
|
Do II równania po wstawieniu wartości teoretycznych za yt1 i yt3 stosujemy zwykłą MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
yt2 = b12 y^t1 + b32 y^t3 + wt2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(b12, b32)' |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
y^t1 |
y^t3 |
|
|
yt2 |
|
|
|
|
|
|
2,5 |
1,5 |
|
|
2 |
|
|
|
|
|
Z= |
0 |
0 |
|
|
0 |
|
8,5 |
|
|
|
|
0 |
0 |
|
y= |
0 |
ZTy= |
4 |
|
|
|
|
-2,5 |
-1,5 |
|
|
-1 |
|
|
|
|
|
|
1 |
-0,5 |
|
|
0 |
|
|
|
|
|
|
-1 |
0,5 |
|
|
-1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
(ZTZ)= |
14,5 |
6,5 |
|
(ZTZ)-1= |
0,165289256198347 |
-0,214876033057851 |
|
|
|
|
|
6,5 |
5 |
|
|
-0,214876033057851 |
0,479338842975207 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
det(ZTZ)= |
30,25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
0,545454545454545 |
b12^ |
oceny parametrów drugiego równania |
|
|
|
|
|
|
|
|
0,090909090909091 |
b32^ |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chcemy policzyć błędy średnie szacunku |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
w. Reszt |
ut2^ =yt2 - b12^ yt1 + b32^ yt3 |
|
|
|
|
|
|
|
|
|
u^= |
0,636363636363637 |
|
|
0,404958677685951 |
|
|
|
|
|
|
|
-0,090909090909091 |
|
u^2= |
0,008264462809917 |
|
|
|
|
|
|
|
-0,454545454545455 |
|
|
0,206611570247934 |
|
|
|
|
|
|
|
0,636363636363636 |
|
|
0,40495867768595 |
|
|
|
|
|
|
|
-0,363636363636364 |
|
|
0,132231404958678 |
|
|
|
|
|
|
|
-0,363636363636364 |
|
|
0,132231404958678 |
|
|
|
|
|
|
|
|
|
|
1,28925619834711 |
suma kw reszt |
|
|
|
|
|
Wariancja resztowa |
|
s2= |
0,322314049586777 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ocena asymptotycznej macierzy kowariancji estymatora 2MNK: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
s2(ZTZ)-1= |
0,053275049518476 |
-0,069257564374018 |
|
|
|
|
|
|
|
|
|
-0,069257564374018 |
0,154497643603579 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Przybliżone błędy średnie szacunku: |
|
|
|
|
|
|
|
|
|
|
D(b12^)= |
0,230813885020974 |
|
|
|
|
|
|
|
|
|
D(b32^)= |
0,393061882664268 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
III równanie |
|
|
|
|
|
|
|
|
|
|
Jest niejednoznacznie identyfikowalne, MNK = 2MNK, bo oderwane |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2MNK do III równania |
|
|
|
|
|
|
|
|
|
|
Równanie jest oderwana 2MNK=MNK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Do III równania stosujemy zwykłą MNK - nie ma zmiennych współzależnych jako zm. objaśniających |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
yt3 = g23 xt2 + ut3 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=g23 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
xt2 |
|
|
yt3 |
|
|
|
|
|
|
|
0 |
|
|
3 |
|
|
|
|
|
|
Z= |
0 |
|
|
1 |
|
|
|
|
|
|
|
0 |
|
y= |
-1 |
ZTy= |
1 |
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
|
|
-1 |
|
|
-2 |
|
|
|
|
|
|
|
1 |
|
|
-1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
(ZTZ)= |
2 |
|
(ZTZ)-1= |
0,5 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
det(ZTZ)= |
2 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a=(ZTZ)-1ZTy= |
|
0,5 |
g23^ |
ocena parametru trzeciego równania |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Chcemy policzyć błędy średnie szacunku |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
w. Reszt |
ut3^= yt3 - g23^ xt2 |
|
|
|
|
|
|
|
|
|
u^= |
3 |
|
u^2= |
9 |
|
|
|
|
|
|
|
1 |
|
|
1 |
|
|
|
|
|
|
|
-1 |
|
|
1 |
|
|
|
|
|
|
|
0 |
|
|
0 |
|
|
|
|
|
|
|
-1,5 |
|
|
2,25 |
|
|
|
|
|
|
|
-1,5 |
|
|
2,25 |
|
|
|
|
|
|
|
|
|
|
15,5 |
suma kw reszt |
|
|
|
|
|
Wariancja resztowa |
|
s2= |
3,1 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Ocena asymptotycznej macierzy kowariancji estymatora 2MNK: |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
s2(ZTZ)-1= |
1,55 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Przybliżony błęd średni szacunku: |
|
|
|
|
|
|
|
|
|
|
D(g23^)= |
1,24498995979887 |
|
|
|
|
|
|
|
|
|