Mean squared error

Residuals Sum of Squares 88.43787 26.05798 Deg. of Freedom 5 66 Somme des carrés inter-classes Residual standard error: 0.6283453 (résiduels) Estimated effects may be unbalanced Ecart-type estimé des résidus XIX- Les Sorties > X2 Pearson's Chi-squared test Valeur de la statistique de test du chi2 data: tab X-squared = 27.1387, df = 1, p-value = 1.894e-07 P(X>X-squared ) X v.a. de loi X²(1) XIX- Les Sorties > st Test t pour variances inégales P(T>t ) Welch Two Sample

Residuals Sum of Squares 88.43787 26.05798 Deg. of Freedom 5 66 Somme des carrés inter-classes Residual standard error: 0.6283453 (résiduels) Estimated effects may be unbalanced Ecart-type estimé des résidus XIX- Les Sorties > X2 Pearson's Chi-squared test Valeur de la statistique de test du chi2 data: tab X-squared = 27.1387, df = 1, p-value = 1.894e-07 P(X>X-squared ) X v.a. de loi X²(1) XIX- Les Sorties > st Test t pour variances inégales P(T>t ) Welch Two Sample

Variable C Rating Thaïlande Taux de change $/euro R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood Durbin-Watson stat 1994:02 2004:02 Coefficient -0.000617 -0.108536 0.053323 0.166388 0.152259 0.004303 0.002185 489.0913 2.002805 Std. Error 0.000397 0.030087 0.014739 121 observations t-Statistic -1.553671 -3.607361 3.617847 Prob. 0.1229 0.0005 0.0004 -0.000417 0.004673 -8.034567 -7.965250 11.77635 0.000022 Mean dependent var S.D. dependent var Akaike info criterion

Résultats de l’estimation du modèle Variable Dépendante: D(LOGINV) Variable Coefficient Std. Error t-Statistic Prob.   COC(-1) -0.013514 0.057008 -0.237063 0.8151 COC(-2) -0.125478 0.054580 -2.298978 0.0330 TXPIB 0.027028 0.008626 3.133235 0.0055 D(LOGEBE) 0.840281 0.220540 3.810115 0.0012 DUMMY -0.106272 0.037229 -2.854574 0.0101 C 0.018976 0.034536 0.549475 0.5891 R-squared 0.655176 Source : Réalisé par les auteurs à partir du logiciel EViews Notons que la DUMMY

comme suit : Lconso = -34,73332*** + 0,202957Lrnbh** + 4,291599Limport*** + 3,171963Lcpop*** Les résultats d’estimation du modèle sont résumés dans le tableau ci-dessous : Variable dépendante : Lconso Variables explicatives Coefficient Std. Error t-Statistic Prob. C -34.73332*** 8.386741 -4.141456 0.0003 Limport 0.202957** 0.057411 3.535163 0.0013 Lrnbh 4.291599*** 0.747236 5.743300 0.0000 Lcpop 3.171963*** 0.455068 6.970297 0.0000 R2 = 0.869068 R2 ajusté= 0.855975 Durbin-Watson stat=

well-being through the reduction of human poverty as measured by the Human Poverty Index (HPI). Thus the aim of this study is to measure the impact of public investments in basic social infrastructure on the reduction of poverty in Benin. Indeed, using an error correction model, a long-term relationship and short-run relationship is estimated to identify long and short term impact of investments in basic social infrastructure on HPI. Analysis of the results reveals that estimates of long-term, only public

applications Arthur CHARPENTIER 2.7.2 Le test du CU SUM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7.3 Le test BDS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7.4 Le test RESE T (Regression error speci…cation test ) . . . . . . . . . . . . . . . 2.7.5 La procédure AUT OREG sous SAS=ET S . . . . . . . . . . . . . . . . . . . . 2.8 Application sur des données réelles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.8.1 Modélisation

============================================================ C 1176.090 207.3921 5.670852 0.0005 REV 0.780983 0.017939 43.53518 0.0000 ============================================================ R–squared 0.995797 Mean dependent var 9985.575 Adjusted R–squared 0.995271 S.D. dependent var 2089.553 S.E. of regression 143.6878 Akaike info criter 10.11214 Sum squared resid 165169.4 Schwartz criterion 10.17266 Log likelihood –62.75009 F–statistic 1895.312 Durbin–Watson stat 1.881665 Prob(F–statistic) 0.000000 ============================================================

|Coefficient |Std. Error |t-Statistic |Prob. | |C |2.459377 |1.637101 |1.502276 |0.1355 | |@TREND |-0.000180 |0.002338 |-0.077175 |0.9386 | |BOUYGUES(-1) |0.960309 |0.026359 |36.43227 |0.0000 | |R-squared |0.918833 | Mean dependent var |60.15789 | |Adjusted R-squared |0.917534

donne le résultat ci-après : Figure 3: estimation naïve du modèle de départ Dependent Variable: LEIB Method: Least Squares Date: 02/10/13 Time: 11:35 Sample (adjusted): 1975 2010 Included observations: 36 after adjustments Variable Coefficient Std. Error t-Statistic Prob. LDCFMEN IPC LEVN LM2 LPIBH TINT RD BP_COURANTS CPIB DEV DLEIB C -7.062121 -0.005776 1.265948 -1.233579 6.639383 0.122177 -0.621598 0.026033 -0.012531 1.278058 0.201003 -53.60436 1.849117 0.006860 1.646017 0.585855 1.751882 0