Analyse des déterminants de l'adoption des semences améliorées au Niger. Cas du mil.

ANNEXES

1. Estimation du modèle

. ^{heckman Rendmttha Prosupficicultiha Utilisatractanimal Exprience Tailm Accengrai, twostep select(Adoption} = ^{Agechef Sexe} > ^{Exprience Lalphachef Tailm Prosupficicultiha Revhormen Accscrdi LAOP dispSem) rhosigma}

^{Heckman selection model} -- ^{two-step estimates Number of obs} = ⁵⁰

^{(regression model with sample selection) Censored obs} = ¹⁵

^{Uncensored obs} = ³⁵

^{Wald chi2(5)} = ^263.70

^Prob > ^chi2 = ^0.0000

			^z	^P>|z|				^Rendmttha ^Prosupficicultiha ^Utilisatractanimal ^Exprience ^Tailm ^Accengrai ^_cons
^1.92271 ^1.112522 ^.1819724 ^-.1785141 ^.2253017 ^-2.974736	^.1276723 ^1.005524 ^.0923456 ^.4126237 ^1.238896 ^2.949352	^15.06 ^1.11 ^1.97 ^-0.43 ^0.18 ^-1.01	^0.000 ^0.269 ^0.049 ^0.665 ^0.856 ^0.313	^1.672477 ^-.8582688 ^.0009782 ^-.9872418 ^-2.202891 ^-8.755359			^Adoption ^Agechef ^Sexe ^Exprience ^Lalphachef ^Tailm ^Prosupficicultiha ^Revhormen ^Accscrdi ^LAOP ^dispSem ^_cons
^Coef. ^.0868776 ^3.692501 ^.0671749 ^.1032777 ^1.619099 ^.1192774 ^-.0001855 ^6.862668 ^3.357243 ^.9469853 ^-13.37616	^Std. Err. ^.0516141 ^2.339588 ^.0717993 ^.7912702 ^.8221708 ^.1686474 ^.0001066 ^3.038351 ^1.682396 ^1.242066 ^6.027023	^1.68 ^1.58 ^0.94 ^0.13 ^1.97 ^0.71 ^-1.74 ^2.26 ^2.00 ^0.76 ^-2.22	^0.092 ^0.115 ^0.349 ^0.896 ^0.049 ^0.479 ^0.082 ^0.024 ^0.046 ^0.446 ^0.026	^[95% Conf. ^-.0142841 ^-.8930061 ^-.0735491 ^-1.447583 ^.0076741 ^-.2112654 ^-.0003944 ^.90761 ^.0598084 ^-1.487419 ^-25.18891			^mills ^lambda
^-.6693853	^1.643981	^-0.41	^0.684	^-3.891529			^rho ^sigma

^-0.24151

^2.7716812

2. Prédiction du modèle

^Probit model
	^D
^Classified	^True ^32 ^3	^~D
. + - ^Total ^Classified ^True D defined	^35 + ^if predicted Pr(D)	^1 ^14 ^15
^Sensitivity ^Specificity ^Positive predictive value ^Negative predictive value		^Pr( +| ^D) ^Pr( -|~D) ^Pr( D| +) ^Pr(~D| -)
^False + ^rate for true ~D ^False + ^rate for classified +		^Pr( +|~D) ^Pr( -| ^D) ^Pr(~D| +) ^Pr( D| -)
^False - ^rate for true D ^Correctly classified

^{Interval] 2.172943 3.083313 .3629665 .6302135 2.653494 2.805887 .1880393 8.278009 .2078989 1.654139 3.230525 .4498202 .0000233 12.81773 6.654678 3.381389 -1.563411 2.552758}

^{for Adoption}

^Total

³³

¹⁷

^{as Adoption} != ⁰

>= .⁵

⁵⁰

^91.43%

^{93.33% 96.97% 82.35% False} - ^{rate for classified} -

^6.67%

^8.57%

^3.03%

^17.65%

^92.00%

DEA-Master/NPTCI Page 33

3. Calcul des effets marginaux

. ^{dprobit Adoption Agechef Sexe Exprience Lalphachef Tailm Prosupficicultiha Revhormen Accscrdi LAOP dispSem}

^{Iteration 0: log likelihood} = ^-30.543215

^{Iteration 1: log likelihood} = ^-15.18848

^{Iteration 2: log likelihood} = ^-12.158841

^{Iteration 3: log likelihood} = ^-10.665289

^{Iteration 4: log likelihood} = ^-9.9596533

^{Iteration 5: log likelihood} = ^-9.6543512

^{Iteration 6: log likelihood} = ^-9.5793765

^{Iteration 7: log likelihood} = ^-9.5770366

^{Iteration 8: log likelihood} = ^-9.5770348

^{Probit regression, reporting marginal effects Number of obs} = ⁵⁰

^{LR chi2(10)} = ^41.93

^Prob > ^chi2 = ^0.0000

^{Log likelihood} = ^{-9.5770348 Pseudo R2} = ^0.6864

^{dF/dx Std. Err. z P>|z| x-bar} [ ^{95% C.I. J}

^Adoption

^Agechef

^Sexe*

^Exprie~e

^Lalpha~f*

^Tailm

^Prosup~a

^Revhor~n

^Accscrdi*

^LAOP*

^dispSem*

^{.0012022 .0037197 1.68 0.092 51.48 -.006088 .008493}

^{.5260741 .3216911 1.58 0.115 .72 -.104429 1.15658}

^{.0009296 .0028064 0.94 0.349 15.04 -.004571 .00643}

^{.0014099 .0134635 0.13 0.896 .44 -.024978 .027798}

^{.0224051 .0640586 1.97 0.049 2.98 -.103147 .147958}

^{.0016506 .0045503 0.71 0.479 7.714 -.007268 .010569}

^{-2.57e-06 6.99e-06 -1.74 0.082 31948 -.000016 .000011}

^{.9905668 .0293363 2.26 0.024 .72 .933069 1.04806}

^{.4300904 .2710561 2.00 0.046 .72 -.10117 .961351}

^{.0217997 .0492427 0.76 0.446 .64 -.074714 .118314}

^{obs. P}

^{pred. P}

^.7

^{.9952406 (at x-bar)}

(*) ^{dF/dx is for discrete change of dummy variable from 0 to 1}

^{z and P>|z| correspond to the test of the underlying coefficient being 0}

.

DEA-Master/NPTCI Page 34