Welcome to SHAZAM - Version 9.0 - OCT 2003 SYSTEM=WIN-98 PAR= 2000 CURRENT WORKING DIRECTORY IS: C:\ESL\USER |_* Exercise 9.17 |_READ (DATA4-3) YEAR HOUSING POP GNP UNEMP INTRATE UNIT 88 IS NOW ASSIGNED TO: DATA4-3 ...SAMPLE RANGE IS NOW SET TO: 1 23 |_GENR LPH=LOG(HOUSING/POP) |_GENR LPCGNP=LOG(GNP/POP) |_GENR LR=LOG(INTRATE) |_OLS LPH LPCGNP LR /DWPVALUE RESID=U REQUIRED MEMORY IS PAR= 8 CURRENT PAR= 2000 OLS ESTIMATION 23 OBSERVATIONS DEPENDENT VARIABLE= LPH ...NOTE..SAMPLE RANGE SET TO: 1, 23 DURBIN-WATSON STATISTIC = 0.80836 DURBIN-WATSON POSITIVE AUTOCORRELATION TEST P-VALUE = 0.000162 NEGATIVE AUTOCORRELATION TEST P-VALUE = 0.999838 R-SQUARE = 0.4812 R-SQUARE ADJUSTED = 0.4294 VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.29171E-01 STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.17079 SUM OF SQUARED ERRORS-SSE= 0.58342 MEAN OF DEPENDENT VARIABLE = 1.9920 LOG OF THE LIKELIHOOD FUNCTION = 9.61944 VARIABLE ESTIMATED STANDARD T-RATIO PARTIAL STANDARDIZED ELASTICITY NAME COEFFICIENT ERROR 20 DF P-VALUE CORR. COEFFICIENT AT MEANS LPCGNP 2.1356 0.6997 3.052 0.006 0.564 1.0805 2.7168 LR -1.1190 0.2729 -4.101 0.001-0.676 -1.4516 -1.2243 CONSTANT -0.98100 1.273 -0.7706 0.450-0.170 0.0000 -0.4925 |_* |_* The Durbin-Watson statistic is strongly significant. |_* Confirm this by running a Breusch-Godfrey test. |_* |_GENR U1=LAG(U) ..NOTE.LAG VALUE IN UNDEFINED OBSERVATIONS SET TO ZERO |_OLS U U1 LPCGNP LR REQUIRED MEMORY IS PAR= 4 CURRENT PAR= 2000 OLS ESTIMATION 23 OBSERVATIONS DEPENDENT VARIABLE= U ...NOTE..SAMPLE RANGE SET TO: 1, 23 R-SQUARE = 0.3176 R-SQUARE ADJUSTED = 0.2099 VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.20953E-01 STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.14475 SUM OF SQUARED ERRORS-SSE= 0.39811 MEAN OF DEPENDENT VARIABLE = -0.30169E-17 LOG OF THE LIKELIHOOD FUNCTION = 14.0144 VARIABLE ESTIMATED STANDARD T-RATIO PARTIAL STANDARDIZED ELASTICITY NAME COEFFICIENT ERROR 19 DF P-VALUE CORR. COEFFICIENT AT MEANS U1 0.58037 0.1952 2.974 0.008 0.564 0.5774*********** LPCGNP -0.39311 0.6075 -0.6470 0.525-0.147 -0.2761*********** LR 0.13829 0.2359 0.5863 0.565 0.133 0.2491*********** CONSTANT 0.69284 1.104 0.6277 0.538 0.143 0.0000*********** |_GEN1 LM=$R2*$N ..NOTE..CURRENT VALUE OF $R2 = 0.31762 ..NOTE..CURRENT VALUE OF $N = 23.000 |_DISTRIB LM /TYPE=CHI DF=1 CHI-SQUARE PARAMETERS- DF= 1.0000 MEAN= 1.0000 VARIANCE= 2.0000 MODE= 0.0000 DATA PDF CDF 1-CDF LM ROW 1 7.3053 0.38262E-02 0.99312 0.68752E-02 |_* |_* The LM test statistic is also significant, indicating the presence of |_* first-order autocorrelation. |_* |_* Since autocorrelation is present, the original estimates, |_* although unbiased and consisitent (there being no lagged dependent |_* variable), are not fully efficient. |_* |_* Reestimate the model using the CORC procedure. |_* |_AUTO LPH LPCGNP LR / RSTAT REQUIRED MEMORY IS PAR= 5 CURRENT PAR= 2000 DEPENDENT VARIABLE = LPH ..NOTE..R-SQUARE,ANOVA,RESIDUALS DONE ON ORIGINAL VARS LEAST SQUARES ESTIMATION 23 OBSERVATIONS BY COCHRANE-ORCUTT TYPE PROCEDURE WITH CONVERGENCE = 0.00100 ITERATION RHO LOG L.F. SSE 1 0.00000 9.61944 0.58342 2 0.55264 14.4325 0.37786 3 0.62606 14.5798 0.37090 4 0.65467 14.5976 0.36933 5 0.66815 14.5982 0.36880 6 0.67500 14.5965 0.36859 7 0.67859 14.5950 0.36849 8 0.68051 14.5941 0.36844 9 0.68155 14.5936 0.36842 10 0.68211 14.5933 0.36840 LOG L.F. = 14.5933 AT RHO = 0.68211 ASYMPTOTIC ASYMPTOTIC ASYMPTOTIC ESTIMATE VARIANCE ST.ERROR T-RATIO RHO 0.68211 0.02325 0.15248 4.47350 R-SQUARE = 0.6724 R-SQUARE ADJUSTED = 0.6397 VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.18420E-01 STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.13572 SUM OF SQUARED ERRORS-SSE= 0.36840 MEAN OF DEPENDENT VARIABLE = 1.9920 LOG OF THE LIKELIHOOD FUNCTION = 14.5933 VARIABLE ESTIMATED STANDARD T-RATIO PARTIAL STANDARDIZED ELASTICITY NAME COEFFICIENT ERROR 20 DF P-VALUE CORR. COEFFICIENT AT MEANS LPCGNP 2.2730 0.7190 3.161 0.005 0.577 1.1500 2.8915 LR -1.4151 0.3029 -4.673 0.000-0.722 -1.8357 -1.5482 CONSTANT -0.67443 1.476 -0.4570 0.653-0.102 0.0000 -0.3386 DURBIN-WATSON = 1.0431 VON NEUMANN RATIO = 1.0905 RHO = 0.45466 RESIDUAL SUM = -0.88244E-01 RESIDUAL VARIANCE = 0.18809E-01 SUM OF ABSOLUTE ERRORS= 2.4537 R-SQUARE BETWEEN OBSERVED AND PREDICTED = 0.6919 RUNS TEST: 8 RUNS, 13 POS, 0 ZERO, 10 NEG NORMAL STATISTIC = -1.8706 DURBIN H STATISTIC (ASYMPTOTIC NORMAL) = 3.1966 MODIFIED FOR AUTO ORDER=1 |_DISTRIB $DURH NORMAL DISTRIBUTION - MEAN= 0.0000 VARIANCE= 1.0000 DATA Z PDF CDF 1-CDF $DURH ROW 1 3.1966 3.1966 0.24098E-02 0.99930 0.69517E-03 |_* |_* A significant value for the Durbin h-statistic indicates that there |_* remains some higher-order autocorrelation unaccounted for. |_* The Durbin-Watson test is biased here, and should not be relied on. |_* |_* Confirm this conclusion using a HILU grid search. |_* |_AUTO LPH LPCGNP LR /GS RSTAT RESID=E REQUIRED MEMORY IS PAR= 5 CURRENT PAR= 2000 DEPENDENT VARIABLE = LPH ..NOTE..R-SQUARE,ANOVA,RESIDUALS DONE ON ORIGINAL VARS LEAST SQUARES ESTIMATION 23 OBSERVATIONS BY GRID SEARCH TO ACCURACY OF .01 ITERATION RHO LOG L.F. SSE 1 -0.90000 -2.56587 1.5660 2 -0.80000 -1.09446 1.4168 3 -0.70000 0.269902 1.2775 4 -0.60000 1.61076 1.1481 5 -0.50000 2.95176 1.0288 6 -0.40000 4.29954 0.91958 7 -0.30000 5.65215 0.82039 8 -0.20000 7.00098 0.73130 9 -0.10000 8.33083 0.65231 10 0.00000 9.61944 0.58342 11 0.10000 10.8372 0.52456 12 0.20000 11.9480 0.47563 13 0.30000 12.9115 0.43639 14 0.40000 13.6872 0.40651 15 0.50000 14.2402 0.38552 16 0.60000 14.5440 0.37289 17 0.70000 14.5788 0.36811 18 0.80000 14.3155 0.37097 19 0.90000 13.6444 0.38249 ITERATION RHO LOG L.F. SSE 20 0.61000 14.5599 0.37206 21 0.62000 14.5731 0.37132 22 0.63000 14.5836 0.37065 23 0.64000 14.5913 0.37006 24 0.65000 14.5963 0.36954 25 0.66000 14.5985 0.36910 26 0.67000 14.5978 0.36874 27 0.68000 14.5944 0.36845 28 0.69000 14.5880 0.36824 29 0.70000 14.5788 0.36811 30 0.71000 14.5666 0.36805 31 0.72000 14.5515 0.36806 32 0.73000 14.5333 0.36816 33 0.74000 14.5121 0.36833 34 0.75000 14.4877 0.36857 35 0.76000 14.4601 0.36889 36 0.77000 14.4291 0.36929 37 0.78000 14.3948 0.36977 38 0.79000 14.3570 0.37033 39 0.71000 14.5666 0.36805 LOG L.F. = 14.5666 AT RHO = 0.71000 ASYMPTOTIC ASYMPTOTIC ASYMPTOTIC ESTIMATE VARIANCE ST.ERROR T-RATIO RHO 0.71000 0.02156 0.14684 4.83532 R-SQUARE = 0.6727 R-SQUARE ADJUSTED = 0.6400 VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.18402E-01 STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.13566 SUM OF SQUARED ERRORS-SSE= 0.36805 MEAN OF DEPENDENT VARIABLE = 1.9920 LOG OF THE LIKELIHOOD FUNCTION = 14.5666 VARIABLE ESTIMATED STANDARD T-RATIO PARTIAL STANDARDIZED ELASTICITY NAME COEFFICIENT ERROR 20 DF P-VALUE CORR. COEFFICIENT AT MEANS LPCGNP 2.2657 0.7247 3.127 0.005 0.573 1.1463 2.8822 LR -1.4401 0.3055 -4.714 0.000-0.725 -1.8683 -1.5757 CONSTANT -0.60051 1.514 -0.3966 0.696-0.088 0.0000 -0.3015 DURBIN-WATSON = 1.0382 VON NEUMANN RATIO = 1.0854 RHO = 0.46136 RESIDUAL SUM = -0.82846E-01 RESIDUAL VARIANCE = 0.18746E-01 SUM OF ABSOLUTE ERRORS= 2.4622 R-SQUARE BETWEEN OBSERVED AND PREDICTED = 0.6969 RUNS TEST: 8 RUNS, 13 POS, 0 ZERO, 10 NEG NORMAL STATISTIC = -1.8706 DURBIN H STATISTIC (ASYMPTOTIC NORMAL) = 3.1163 MODIFIED FOR AUTO ORDER=1 |_* |_* Do a Breusch-Godfrey LM test. |_GENR ELAG=LAG(E) ..NOTE.LAG VALUE IN UNDEFINED OBSERVATIONS SET TO ZERO |_OLS E ELAG LPCGNP LR REQUIRED MEMORY IS PAR= 5 CURRENT PAR= 2000 OLS ESTIMATION 23 OBSERVATIONS DEPENDENT VARIABLE= E ...NOTE..SAMPLE RANGE SET TO: 1, 23 R-SQUARE = 0.2746 R-SQUARE ADJUSTED = 0.1600 VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.14303E-01 STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.11960 SUM OF SQUARED ERRORS-SSE= 0.27176 MEAN OF DEPENDENT VARIABLE = -0.36020E-02 LOG OF THE LIKELIHOOD FUNCTION = 18.4051 VARIABLE ESTIMATED STANDARD T-RATIO PARTIAL STANDARDIZED ELASTICITY NAME COEFFICIENT ERROR 19 DF P-VALUE CORR. COEFFICIENT AT MEANS ELAG 0.45606 0.2058 2.216 0.039 0.453 0.4479 -0.1854 LPCGNP -0.56494 0.4932 -1.145 0.266-0.254 -0.4952 397.4340 LR 0.25619 0.1911 1.341 0.196 0.294 0.5758 -155.0095 CONSTANT 0.86894 0.9033 0.9620 0.348 0.216 0.0000 -241.2391 |_GEN1 LM=$R2*$N ..NOTE..CURRENT VALUE OF $R2 = 0.27455 ..NOTE..CURRENT VALUE OF $N = 23.000 |_DISTRIB LM /TYPE=CHI DF=1 CHI-SQUARE PARAMETERS- DF= 1.0000 MEAN= 1.0000 VARIANCE= 2.0000 MODE= 0.0000 DATA PDF CDF 1-CDF LM ROW 1 6.3147 0.67534E-02 0.98803 0.11974E-01 |_* |_* Since there appears to be additional autocorrelation, test the |_* original residuals for an AR(3) process. |_* |_GENR U2=LAG(U1) ..NOTE.LAG VALUE IN UNDEFINED OBSERVATIONS SET TO ZERO |_GENR U3=LAG(U2) ..NOTE.LAG VALUE IN UNDEFINED OBSERVATIONS SET TO ZERO |_OLS U U1-U3 LPCGNP LR REQUIRED MEMORY IS PAR= 6 CURRENT PAR= 2000 OLS ESTIMATION 23 OBSERVATIONS DEPENDENT VARIABLE= U ...NOTE..SAMPLE RANGE SET TO: 1, 23 R-SQUARE = 0.6451 R-SQUARE ADJUSTED = 0.5407 VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.12181E-01 STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.11037 SUM OF SQUARED ERRORS-SSE= 0.20707 MEAN OF DEPENDENT VARIABLE = -0.30169E-17 LOG OF THE LIKELIHOOD FUNCTION = 21.5317 VARIABLE ESTIMATED STANDARD T-RATIO PARTIAL STANDARDIZED ELASTICITY NAME COEFFICIENT ERROR 17 DF P-VALUE CORR. COEFFICIENT AT MEANS U1 0.83026 0.2309 3.595 0.002 0.657 0.8260*********** U2 -0.47892 0.2788 -1.718 0.104-0.385 -0.4753*********** U3 -0.22198 0.2410 -0.9212 0.370-0.218 -0.2145*********** LPCGNP -0.52931 0.4702 -1.126 0.276-0.263 -0.3718*********** LR 0.18499 0.1816 1.019 0.323 0.240 0.3332*********** CONSTANT 0.93437 0.8546 1.093 0.289 0.256 0.0000*********** |_GEN1 LM=$R2*$N ..NOTE..CURRENT VALUE OF $R2 = 0.64507 ..NOTE..CURRENT VALUE OF $N = 23.000 |_DISTRIB LM /TYPE=CHI DF=3 CHI-SQUARE PARAMETERS- DF= 3.0000 MEAN= 3.0000 VARIANCE= 6.0000 MODE= 1.0000 DATA PDF CDF 1-CDF LM ROW 1 14.837 0.92221E-03 0.99804 0.19616E-02 |_* |_* The LM test statistic is significant. However, the t-statistics suggest |_* that the third-order lag is not significant, which is consistent |_* with an AR(2) process. |_* |_* Estimate the AR(3) model. |_* |_AUTO LPH LPCGNP LR / ORDER=3 DEPENDENT VARIABLE = LPH ..NOTE..R-SQUARE,ANOVA,RESIDUALS DONE ON ORIGINAL VARS REQUIRED MEMORY IS PAR= 8 CURRENT PAR= 2000 AUTOREGRESSIVE ERROR MODEL, ORDER= 3 ITERATION 0 ESTIMATES AND ERROR SUM OF SQUARES 2.1356 -1.1190 -0.98100 0.0000 0.0000 0.0000 0.58342 ITERATION 1 ESTIMATES AND ERROR SUM OF SQUARES 1.6063 -0.93400 -0.46633E-01 -0.83026 0.47892 0.22198 0.22993 ITERATION 2 ESTIMATES AND ERROR SUM OF SQUARES 2.0798 -1.1557 -0.75796 -0.83408 0.52616 0.14512 0.21767 ITERATION 3 ESTIMATES AND ERROR SUM OF SQUARES 2.0219 -1.1689 -0.57950 -0.94871 0.58937 0.11168 0.21560 ITERATION 4 ESTIMATES AND ERROR SUM OF SQUARES 2.1328 -1.2288 -0.72832 -0.94645 0.58847 0.89622E-01 0.21473 ITERATION 5 ESTIMATES AND ERROR SUM OF SQUARES 2.0691 -1.2289 -0.56515 -1.0096 0.62031 0.66814E-01 0.21421 ITERATION 6 ESTIMATES AND ERROR SUM OF SQUARES 2.1553 -1.2776 -0.67508 -1.0146 0.62914 0.44972E-01 0.21372 ITERATION 7 ESTIMATES AND ERROR SUM OF SQUARES 2.0764 -1.2795 -0.46884 -1.0973 0.67651 0.14370E-01 0.21309 ITERATION 8 ESTIMATES AND ERROR SUM OF SQUARES 2.2200 -1.3697 -0.63206 -1.1103 0.69393 -0.24812E-01 0.21181 ITERATION 9 ESTIMATES AND ERROR SUM OF SQUARES 2.0603 -1.3779 -0.20488 -1.2816 0.80517 -0.87489E-01 0.20829 ITERATION 10 ESTIMATES AND ERROR SUM OF SQUARES 2.4444 -1.6730 -0.51900 -1.3554 0.88569 -0.20199 0.19669 ITERATION 11 ESTIMATES AND ERROR SUM OF SQUARES 2.2517 -1.7152 0.64614E-01 -1.6111 1.1018 -0.26490 0.17976 ITERATION 12 ESTIMATES AND ERROR SUM OF SQUARES 2.6053 -2.0046 -0.19688 -1.6648 1.1511 -0.30684 0.16912 ITERATION 13 ESTIMATES AND ERROR SUM OF SQUARES 2.5132 -2.0127 0.36754E-01 -1.7673 1.2458 -0.31001 0.16557 ITERATION 14 ESTIMATES AND ERROR SUM OF SQUARES 2.6074 -2.0924 -0.28323E-01 -1.7711 1.2404 -0.30994 0.16463 ITERATION 15 ESTIMATES AND ERROR SUM OF SQUARES 2.5820 -2.0947 0.34898E-01 -1.7974 1.2672 -0.31360 0.16441 ITERATION 16 ESTIMATES AND ERROR SUM OF SQUARES 2.6005 -2.1117 0.24735E-01 -1.7976 1.2648 -0.31300 0.16437 ITERATION 17 ESTIMATES AND ERROR SUM OF SQUARES 2.5948 -2.1122 0.38741E-01 -1.8033 1.2708 -0.31404 0.16436 ITERATION 18 ESTIMATES AND ERROR SUM OF SQUARES 2.5985 -2.1157 0.37066E-01 -1.8033 1.2702 -0.31389 0.16436 ITERATION 19 ESTIMATES AND ERROR SUM OF SQUARES 2.5973 -2.1158 0.40012E-01 -1.8045 1.2715 -0.31412 0.16436 RESIDUAL CORRELOGRAM LM-TEST FOR HJ:RHO (J)=0,STATISTIC IS CHI-SQUARE(1) LAG RHO STD ERR T-STAT LM-STAT 1 -0.0038 0.2085 -0.0180 0.0040 2 -0.0178 0.2085 -0.0852 0.0133 3 0.0986 0.2085 0.4727 0.3941 4 -0.1490 0.2085 -0.7145 0.8800 5 0.0858 0.2085 0.4117 0.3084 6 0.0633 0.2085 0.3035 0.1522 CHISQUARE WITH 6 D.F. IS 1.003 ASYMPTOTIC ESTIMATE VARIANCE ST.ERROR T-RATIO RHO 1 1.80451 0.04091 0.20226 8.92177 RHO 2 -1.27149 0.12070 0.34742 -3.65981 RHO 3 0.31412 0.04780 0.21864 1.43669 R-SQUARE = 0.8539 R-SQUARE ADJUSTED = 0.8392 VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.82179E-02 STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.90652E-01 SUM OF SQUARED ERRORS-SSE= 0.16436 MEAN OF DEPENDENT VARIABLE = 1.9920 VARIABLE ESTIMATED STANDARD T-RATIO PARTIAL STANDARDIZED ELASTICITY NAME COEFFICIENT ERROR 20 DF P-VALUE CORR. COEFFICIENT AT MEANS LPCGNP 2.5973 0.4893 5.308 0.000 0.765 1.3140 3.3041 LR -2.1158 0.2408 -8.786 0.000-0.891 -2.7448 -2.3149 CONSTANT 0.40012E-01 0.9801 0.4082E-01 0.968 0.009 0.0000 0.0201 |_* |_* The estimate of Rho3 is not significantly different from zero. |_* Reestimate using an AR(2) specification. |_* |_AUTO LPH LPCGNP LR / ORDER=2 REQUIRED MEMORY IS PAR= 6 CURRENT PAR= 2000 DEPENDENT VARIABLE = LPH ..NOTE..R-SQUARE,ANOVA,RESIDUALS DONE ON ORIGINAL VARS LEAST SQUARES SECOND-ORDER AUTOCORRELATION BY COCHRANE-ORCUTT TYPE PROCEDURE WITH CONVERGENCE =0.001000 23 OBSERVATIONS ITERATION RHO1 RHO2 SSE SSE/N LOG.L.F. 1 0.00000 0.00000 0.58341546 0.25365890E-01 9.6194379 2 1.01958 -0.72282 0.15987600 0.69511303E-02 23.551512 3 1.12698 -0.75285 0.14958317 0.65036162E-02 24.168286 4 1.18026 -0.76792 0.14603642 0.63494094E-02 24.361562 5 1.20868 -0.77559 0.14460098 0.62869990E-02 24.429662 6 1.22444 -0.77962 0.14395554 0.62589366E-02 24.455843 7 1.23341 -0.78181 0.14363917 0.62451814E-02 24.466835 8 1.23861 -0.78305 0.14347347 0.62379768E-02 24.471860 9 1.24166 -0.78377 0.14338250 0.62340218E-02 24.474337 10 1.24346 -0.78418 0.14333098 0.62317819E-02 24.475634 11 1.24452 -0.78443 0.14330122 0.62304877E-02 24.476345 12 1.24515 -0.78457 0.14328380 0.62297306E-02 24.476747 ASYMPTOTIC ASYMPTOTIC ASYMPTOTIC ESTIMATE VARIANCE ST.ERROR T-RATIO AUTOCORRELATION RHO1 1.24515 0.01672 0.12929 9.63093 0.69773 RHO2 -0.78457 0.01672 0.12929 -6.06845 0.08421 COVARIANCE -0.01166 COMPLEX ROOTS - AUTOREGRESSIVE PROCESS DISPLAYS PSEUDO PERIODIC BEHAVIOUR WITH DAMPED SINE WAVE R-SQUARE = 0.8726 R-SQUARE ADJUSTED = 0.8599 VARIANCE OF THE ESTIMATE-SIGMA**2 = 0.71642E-02 STANDARD ERROR OF THE ESTIMATE-SIGMA = 0.84642E-01 SUM OF SQUARED ERRORS-SSE= 0.14328 MEAN OF DEPENDENT VARIABLE = 1.9920 LOG OF THE LIKELIHOOD FUNCTION = 24.4767 VARIABLE ESTIMATED STANDARD T-RATIO PARTIAL STANDARDIZED ELASTICITY NAME COEFFICIENT ERROR 20 DF P-VALUE CORR. COEFFICIENT AT MEANS LPCGNP 3.3510 0.5944 5.638 0.000 0.783 1.6954 4.2629 LR -1.7181 0.2375 -7.233 0.000-0.851 -2.2289 -1.8798 CONSTANT -2.7582 1.097 -2.513 0.021-0.490 0.0000 -1.3847 |_* |_* We conclude from these results that the elasticity of per capita |_* housing with respect to percapita GNP is relatively high, at 3.35. |_* The interest rate elasticity is -1.72, which is also fairly high. |_* The OLS estimates substantially underestimate both effects. |_* These results suggest that housing is highly sensitive to both |_* economic activity and interest rates. |_* |_* However, note that there is no price index for housing in the model. |_* Therefore, the estimates msy suffer from omitted variable bias. ..INPUT FILE COMPLETED..TYPE A NEW COMMAND OR TYPE: STOP