Heterocedasticidad

EJEMPLO DE HETEROCEDASTICIDAD

Una empresa genera un cash-flow anual (y); los gastos comerciales (x1) y el nivel de inventarios (x2) se muestran en el cuadro:

|obs |Y |X1 |X2 |
|1974 | 972 | 30 | 50 |
|1975 | 1065 | 30 | 89 ||1976 | 1138 | 31 | 116 |
|1977 | 1224 | 32 | 135 |
|1978 | 1181 | 33 | 160 |
|1979 | 1230 | 34 | 177 |
|1980 | 1327 | 36 | 188 |
|1981 | 1357| 38 | 191 |
|1982 | 1344 | 38 | 197 |
|1983 | 1380 | 40 | 211 |
|1984 | 1728 | 46 | 238 |
|1985 | 1797 | 49 | 251 |
|1986 | 1873| 56 | 263 |
|1987 | 2056 | 61 | 266 |
|1988 | 2287 | 63 | 264 |
|1989 | 2188 | 68 | 262 |
|1990 | 2213 | 75 | 263 |
|1991 | 2356 | 80| 258 |
|1992 | 2690 | 86 | 238 |
|1993 | 2919 | 90 | 227 |

Se estimó un modelo de regresión para explicar su flujo de efectivo, los resultados son:

|Dependent Variable: Y |
|Method: LeastSquares |
|Sample: 1974 1993 |
|Included observations: 20 |
|Variable |Coefficient |Std. Error |t-Statistic |Prob. |
|C|187.2384 |73.72084 |2.539830 |0.0211 |
|X1 |25.68271 |1.596001 |16.09191 |0.0000 |
|X2 |1.101824 |0.501250 |2.198152 |0.0421 |
|R-squared |0.974883 | Mean dependent var |1716.250 |
|AdjustedR-squared |0.971928 | S.D. dependent var |578.3670 |
|S.E. of regression |96.90379 | Akaike info criterion |12.12280 |
|Sum squared resid |159635.8 | Schwarz criterion |12.27216 |
|Log likelihood |-118.2280 | F-statistic |329.9147 |
|Durbin-Watsonstat |1.383965 | Prob(F-statistic) |0.000000 |

Se le realizaron pruebas de heterocedasticidad:
1) Residuales al cuadrado con respecto a los valores estimados:

[pic]

2) Prueba de Park
|Dependent Variable: LOG(U2) |
|Variable |Coefficient |Std. Error|t-Statistic |Prob. |
|C |-3.226402 |4.315579 |-0.747617 |0.4643 |
|LOG(X1) |2.773047 |1.113333 |2.490761 |0.0227 |
|Dependent Variable: LOG(U2) |
|Variable |Coefficient |Std. Error...