정책연구시리즈 2009-17 비시장재가치측정에관한연구 - 이중경계양분선택형 CVM 조사의제시금액분석을중심으로 - 김강수
2009 년도정책연구시리즈 비시장재가치측정에관한 연구 - 이중경계양분선택형 CVM 조사의제시금액분석을중심으로 -
발간사.. CVM.,,., CVM..,. 2009 12
목 차 발간사 요약 1 제 1 장연구의배경및목적 5 제 2 장제시금액산출방법에대한기존연구고찰 8 제 3 장예비타당성조사에서의 CVM 조사및제시금액 14 제 1 절예비타당성조사에서의 CVM 조사절차 14 제 2 절예비타당성조사에서의제시금액및응답비율 18 제4장모의실험을통한제시금액분석 24 제1절개요 24 제2절자료생성을위한모의실험 25 제3절제시금액에따른지불의사액추정결과 31 제 5 장종합및결론 40 참고문헌 46
표목차 < 표 2-1> 확률분포별최적제시금액 11 < 표 3-1> 응답형태별비율 21 < 표 3-2> A사업최초제시금액별 응답유형별답변비율 22 < 표 3-3> B사업최초제시금액별 응답유형별답변비율 23 < 표 4-1> 분석시나리오및제시금액 ( 참지불의사액 = 3,000원 ) 32 < 표 4-2> 제시금액별분석결과 ( 참지불의사액 = 3,000원 ) 33 < 표 4-3> 제시금액수에따른추정지불의사액편의와분산 ( 참지불의사액 = 3,000원, s= 0.5) 34 < 표 4-4> 제시금액산출방법별지불의사액편의와분산 36 < 표 4-5> 사전조사의정확성에따른제시금액방법별비교 ( 참지불의사액 = 3,000원, s= 0.5를가정 ) 38 < 표 4-6> 응답형태별비율 ( 참지불의사액 = 3,000원, s= 0.5) 39 그림목차 [ 그림 4-1] 로그-로지스틱확률밀도함수 26 [ 그림 4-2] 제시금액수에따른추정지불의사액편의와분산 35 [ 그림 4-3] 제시방법별모의실험결과 ( 참지불의사액 = 3,000원, s= 0.1, 0.5, 1.0) 36
요약 (CVM).,. CVM...,,.., no-no. ( 2),
.,.,.,.., 1 2.,, 1.,.,.,,, 1., 1 2.
.,., (, 1,000 300 ),,..,.. no-no. - no-no 56%.,, CV..,.
제 1 장연구의배경및목적 (private goods) (non-marketed goods) (economic values). (Contingent Valuation Method, CVM), (Willness to Pay, WTP). CVM,,,.,., 1,000 8,000 WTP
(asymptotic variance),. CVM.,.,. (WTP),,.. 1 2.,. 3 CVM., 4. 4.,,.
..,. 3~4 CVM.
제 2 장제시금액산출방법에대한기존연구고찰 CVM.,. Hanemann(1984) (Hanemann et al.[1991]). McFadden(1994). Bateman et al.(2000) 1). McConnell(1990) WTP., 1) (2003).
. 2)., ( )., ( / ), 1, 2, 3). 4), (optimal bid design) (Finney[1978]; Abdelbasit and Plackett[1983]; Kanninen[1993a, 1993b]). 5) Ainger(1979), Minkin(1987), Abdelbasit and Plackett(1983), Wu(1988), Kanninen(1993a)., Kanninen(1993a) (Fisher information) (determinant) 2) 20 CVM 17. 17 2. 3) WTP, (, ). 4),, -. 5) (biostatistic).
6). (asymptotic variance). 7), Kanninen(1993a),. (1),,, (distance)., Kanninen(1993a) C-., 8), C-, (delta). 9),, (0.25, 0.50, 0.75). Wu(1988). (2) Alberini(1995) D- C- 6) D-. 7). 8). 9) Wu(1988),.
< 표 2-1> 확률분포별최적제시금액 D- C- exp exp :.,.,.. < 2-1> Kanninen (1993a) Alberini(1995), Kanninen(1993a) D- 12.1%., 87.8%,.,.,, CVM. 10) 10), 5~10..
,,,. (sequential design procedure) (Robbins and Monro[1958]; Dixon and Mood[1948]; Wu[1985]). 11) Robbins and Monro(1958) Dixon and Mood(1948) (non-parametric), Wu(1985) (parametric),. Kanninen(1993b) CVM,., Kanninen(1993b) ( ) C-, ( ) ( ). argmin (3) 11),.
, C-., CVM 3 1, 2, 3 1,, 2, 1 2 3. 1, 2, 3.,,..,,.
제 3 장예비타당성조사에서의 CVM 조사및제시금액 제 1 절예비타당성조사에서의 CVM 조사절차 CVM, CVM. CVM. 12),,,.,. CVM., (CV), 13). 12) ( 5 ) (, 2009), CVM. 13) 6~12 1 2.
., 25~100 100. (open ended questions) 14)., 5.. ( ) 5? ( ),. 0 ( 5 ) (, 2009) Hanemann and Kanninen(1999), (pre-test) WTP CV., WTP 15~85% 4~6. 4~6,.,,,,. 14)..
(dichotomous choice) 15)., ( 2 ), ( 1/2)., (oneand-a-half bounded),,,. Box., CV /.,,,,,,, 0, -. CV,.., CV, 15) Bishop and Herberlein(1979) CVM.,,.
7. ( ) 5 [ ]( )? 7-1 7-2 7-1., ( ) 5 [ 2 ] ( )? 8 8 7-2., ( ) 5 [ 1/2 ]( )? 8 7-3 7-3., ( )?. 7-4. 8,, CV.
제 2 절예비타당성조사에서의제시금액및응답비율 1. 제시금액., 0, 5~10. ( 5 ) (, 2009) (pre-test), 15~85% 4~6. 2009 1 17., 0., 1 500 9, 1,000 7 500 1,000. 17 8 10,000, 30,000.,, 16) 1% 28%. 500 1,500 16) 15%, 85%.
.,, 37%.,. 2. 응답비율 17).,,,, CV. CVM. 18). 17) 17. 18) CVM, CVM.,,,,.,,, WTP. (2009) CVM,,,, /,, /.
< 3-1> 17,., - yes-yes 16%, - no-no 56% yes-yes no-no 72.0%., no-no, CV,., no-no no-no 19), 0 0 (protestzero bids) (protest-bids). 20) 0 (1999)., 1, 2, 0 30%., Halstead et al.(1992),.,, (self-selection bias). 0., 19) 0. 20) Halstead et al.(1992).
< 표 3-1> 응답형태별비율 yes/yes yes/no no/yes no/no yes/yes + no/no 1,000 160 16.0% 171 17.1% 109 10.9% 560 56.0% 720 72.0%, (, 0 ) (, ). no-no,. 21) no-no (2009) 15%,,., A, B.. 21) 17 6,.
< 표 3-2> A 사업최초제시금액별 응답유형별답변비율 1,000 2,000 3,000 4,000 5,000 6,000 7,000 1,000 143 143 143 143 143 143 142 yes/yes yes/no no/yes no/no yes/yes + no/no 109 48 24 8 10 10 5 4 10.90% 33.57% 16.78% 5.59% 6.99% 6.99% 3.50% 2.82% 159 34 27 35 19 16 14 14 15.90% 23.78% 18.88% 24.48% 13.29% 11.19% 9.79% 9.86% 99 9 20 16 22 11 15 6 9.90% 6.29% 13.99% 11.19% 15.38% 7.69% 10.49% 4.23% 633 52 72 84 92 106 109 118 63.30% 36.36% 50.35% 58.74% 64.34% 74.13% 76.22% 83.10% 742 100 96 92 102 116 114 122 74.20% 69.93% 67.13% 64.34% 71.33% 81.12% 79.72% 85.92% WTP 15~85% 7, 2 1/2 2. A 7,000 22) yes-yes 2.82%, 5,000 15%., 1,000 23) no-no 36.36%. 15% 2,,, no-no, 1,000. 22) 7,000 3,500 14,000. 23) 1,000 500 2,000.
< 표 3-3> B 사업최초제시금액별 응답유형별답변비율 1,000 2,000 3,000 4,000 5,000 6,000 7,000 1,000 143 143 143 143 143 143 142 yes/yes yes/no no/yes no/no yes/yes + no/no 100 46 21 8 6 10 5 4 10.00% 32.17% 14.69% 5.59% 4.20% 6.99% 3.50% 2.82% 158 30 29 21 19 23 22 14 15.80% 20.98% 20.28% 14.69% 13.29% 16.08% 15.38% 9.86% 99 6 12 21 14 8 19 19 9.90% 4.20% 8.39% 14.69% 9.79% 5.59% 13.29% 13.38% 643 61 81 93 104 102 97 105 64.30% 42.66% 56.64% 65.03% 72.73% 71.33% 67.83% 73.94% 743 107 102 101 110 112 102 109 74.30% 74.83% 71.33% 70.63% 76.92% 78.32% 71.33% 76.76% B A. 7,000 yes-yes 2.82%, 5,000 15%. 1,000 no-no 42.66%.,. WTP, 2 1/2 2,,.
제 4 장모의실험을통한제시금액분석 제 1 절개요.. ( 2 ),. ( 1 ), 2 1/2 2. 1 2,.
제 2 절자료생성을위한모의실험 1. 모의실험 (WTP) -.,. 24) ( STEP). STEP 1: log(wtp) n(n=1,000). log(wtp), - ( ) ( ).,.,. [ 4-1] (s), 3,000. 25) STEP 2:. STEP 1, 100. 24) -,.,. 25) 3,000.,. ( ) ( ) ( ), (X) (, c ),.
[ 그림 4-1] 로그 - 로지스틱확률밀도함수 STEP 3:., 100, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% 26) 2., 2 1/2 2., 26) (, 2009) 15~85%, WTP 10~90%. WTP 2 15% 85%. WTP,. 10~90%, 15~85% 4~6., 500.
100, 27). STEP 4:. STEP 1 1,000 STEP 3 2., 28) STEP 3, 2, -. STEP 6:.. STEP 7: STEP 1 STEP 6 200., (bias) (variance). 참지불의사액 (4) (5), 27) -. 28) 3,000.
2. 제시금액산정가. 최적제시금액 : 제시금액방법 1., -,. log (6) ( WTP). - log(wtp),. (7)., log(wtp), D-. log log (8) 중앙값 exp exp, exp (9).,.
나. 일정범위 : 제시금액방법 2 ( 5 ) (, 2009), 29), (pre-test)., 15~85% 4~6. 30) 100, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 2., 2 1/2 2. 3. 이중경계양분선택형모형추정및응답자료 (double bound dichotomous choice: DBDC).,.. 29) Hanemann and Kanninen(1999), 82.4% 17.6%, 87.2% 12.75%., Kanninen(1993a), 87.8% 12.1%. 30) (2009),.
. -, 2. 2 4, 2 (yy); 2 (nn); (yn); (ny)., 31) 2 - (nn)., (, 2 2 ) (, 2 1/2 )., or,., 2 (, 1:, 0: ). P r P r P r P r P r Pr P r P r Pr P r P r P r (10) (11) (12) (13),. 31) 3,000.
ln ln ln ln ln (14) (14a) (14b) (14c) (14d), -. 제 3 절제시금액에따른지불의사액추정결과 1. 최초제시금액의범위및수에따른추정지불의사액 ( 2)., 10~90% 15~85% 5~10.,.. 5 32) 32) 5, 3.
< 표 4-1> 분석시나리오및제시금액 ( 참지불의사액 =3,000 원 ) ( ) 1 1000(10%), 3000, 5000, 7000, 9000(90%) 2 1000(10%), 2500, 4000, 5500, 7000(84.5%) 3 1000(10%), 2000, 3000, 4000, 5000(73.5%) 4 1000(10%), 1500, 2000 2500, 3000(50%) 5 2000(30.8%), 3250, 5500, 7750, 9000(90%) 6 3000(50%), 4500, 6000, 7500, 9000(90%) 7 4000(64%), 5250, 6500, 8750, 9000(90%) : ( )., 33) 2 1/2 2. 1~4 1,000, 5~7., 7 3,000, 6. < 4-2>,., 3, 1,000~5,000,.. 33) 1,000., 3,000.
< 표 4-2> 제시금액별분석결과 ( 참지불의사액 =3,000 원 ) ( ) (%) no-no no-yes yes-no yes-yes 1000, 3000, 5000, 7000, 9000 2993.19 93.72 0.38 0.17 0.23 0.21 1000, 2500, 4000, 5500, 7000 2994.89 90.02 0.30 0.21 0.25 0.24 1000, 2000, 3000, 4000, 5000 2996.32 84.61 0.21 0.25 0.25 0.29 1000, 1500, 2000 2500, 3000 2994.71 89.95 0.11 0.29 0.20 0.40 2000, 3250, 5500, 7750, 9000 2993.76 92.28 0.42 0.19 0.26 0.13 3000, 4500, 6000, 7500, 9000 2995.53 94.21 0.47 0.16 0.28 0.09 4000, 5250, 6500, 8750, 9000 2995.34 97.23 0.53 0.14 0.27 0.06,. 1(1,000, 3000, 5,000, 7,000, 9,000) 7(4,000, 5,250, 6,500, 8,750, 9,000), 7. 6 3..,., no-no, no-no no-no. no-no,
CV.., 10%, 90%, 3~7. < 4-3> 10~90%.,. 10%, 50%, 90%,. 7 3,.,. < 표 4-3> 제시금액수에따른추정지불의사액편의와분산 ( 참지불의사액 =3,000원, s=0.5) (%) 3 10, 50, 90 3000.70 96.03 4 10, 36.7, 63.3, 90 2997.72 85.90 5 10, 30, 50, 70, 90 3003.92 88.79 6 10, 26, 42, 58, 74, 90 3002.38 88.68 7 10, 23.3, 36.7, 50, 63.3, 76.7, 90 3002.13 88.57
[ 그림 4-2] 제시금액수에따른추정지불의사액편의와분산 2. 제시금액산출방법에따른추정지불의사액., 34) 1 2., 1 2,. 가. 사전조사의신뢰성에따른비교 < 4-4> 34),.,.,.
. 3,000,., < 표 4-4> 제시금액산출방법별지불의사액편의와분산 s 1 2 3000 0.1-1.86 298.61-3.26 418.05 3000 0.5-8.72 8108.13-11.03 8347.37 3000 1 22.55 32462.89 21.71 36264.16 [ 그림 4-3] 제시방법별모의실험결과 ( 참지불의사액 =3,000 원, s=0.1, 0.5, 1.0)
., 1 2, 1, 2 1.,. 나. 사전조사의정확성에따른비교., (3,000 ) (6,000 ) (1,500 ),., (0.5) (0.7) (0.3),. < 4-5> 1 2., (3000, 0.5), 1 2., 1.,, 1,500 3,000,
< 표 4-5> 사전조사의정확성에따른제시금액방법별비교 ( 참지불의사액 =3,000 원, s=0.5 를가정 ) median s 1 2 1500 0.3 0.47 9839 4.36 8995 1500 0.5 1.13 8750 1.89 10864 1500 0.7-8.52 8919-2.62 10109 3000 0.3-12.17 7439-11.95 7673 3000 0.5 5.72 7808 1.31 8514 3000 0.7 6.04 7920 6.70 9051 6000 0.3-6.25 9689-3.84 8837 6000 0.5-2.21 8298-3.53 9561 6000 0.7-5.27 9034-7.43 9347 1 2..,, 1 2.,. < 4-6>,, 2 no-no 49%, 1 20%.,
< 표 4-6> 응답형태별비율 ( 참지불의사액 =3,000 원, s=0.5) median s 1500 0.3 3000 0.5 6000 0.7 nn ny yn yy 1 0.07 0.13 0.25 0.55 2 0.08 0.15 0.27 0.50 1 0.13 0.38 0.37 0.12 2 0.28 0.23 0.22 0.27 1 0.20 0.59 0.19 0.02 2 0.49 0.19 0.15 0.17 2, 2 1 yes 1 2 no 1/2,.
제 5 장종합및결론 (CVM).,. CVM... 2009 1 17,, 1% 28%, 37%.
,.., - yes-yes 16%, - no-no 56% yes-yes no-no 72.0%., no-no., CV. ( 2),.,.,..,, no-no. no-no.,.,.
,., 3 ~7,.,.., 1 2.,, 1.,.,..,,, 1.,, 1,.,
1 2. CVM., 2,.,,.,.,.,,,,..,., (, 1,000 300 ),
..,, 1.,...,...,,,.. -.,
.,.. no-no. - no-no 56%.,, CV..,.
참고문헌,,, 10 1, 2003., ( ) :,, 8 1, 1999., ( 5 ), 2009. Abdelbasit, K. M. and R. L. Plackett, Experimental Design for Binary Data, Journal of the American Statistical Association, No. 78, 1983, pp.90~98. Ainger, D. J., A Brief Introduction to the Methodology of Optimal Experimental Design, Journal of Econometrics, No. 11, 1979, pp.7~16. Alberini, A., Optimal Designs for Discrete Choice Contingent Valuation Surveys: Single-Bound, Double-Bound, and Bivariate Models, Journal of Environmental Economics and Management, Vol. 28, No. 3, 1995, pp.287~306. Alberini, A. and R. T. Carson, Efficient Threshold Values or Binary Discrete Choice Contingent Valuation Surveys and Economic Experiments, Working paper, Department of Economics, University of California, San Diego, 1990. Bateman, I., I. Langford, and G. Kerr, Bound and Path Effects in Double and Triple Bounded Dichotomous Choice Contingent Valuation, Paper Presented at 10th Annual Conference of the Euripean Association of Environmental and Resources Economists (EAERE), Rethymno, Greece, 2000. Bishop R. C. and T. A. Herberlein, Measuring Values of Extra-Market Goods: Are Indirect Measures Biased? American Journal Agricultural Economics, No. 61, 1979, pp.926~930. Dixon, W. and A. Mood, A Method for Obtaining and Analysing Sensitivity Data, Journal of America Statistics, Vol. 43, 1948, pp.109~126. Finney, D., Statistical Methods in Biological Assays, Oxford University Press, New York, 1978. Halstead, J. M., A. E. Luloff, and T. H. Stevens, Protest Bidders in Contigent
Valuation, Northeastern Journal of Agricultural and Resource Economics, Northeastern Agricultural and Resource Economics Association, Springer, Vol. 1, No. 2, 1992, pp.147~184. Hanemann, M., Welfare Evaluations in Contingent Valuation Experiments with Discrete Responses, American Journal of Agricultural Economics, No. 14, 1984. Hanemann, M. and B. Kanninen, The Statistical Analysis of Discrete-Response CV Data, in I. Bateman and K. C. Willis (eds.), Valuing Environmental Preferences: Theory and Practice of the Contingent Valuation Method on US. EC and Developing Countries, Oxford University Press, 1999. Hanemann, M., J. B. Loomis, and B. J. Kanninen, Statistical Efficiency of Double- Bounded Dichotomous Choice Contingent Valuation, American Journal of Agricultural Economics, No. 73, 1991. Kanninen B. J., Optimal Experimental Design for Double-Bounded Dichotomous Choice Contingent Valuation, Land Economics, No. 69, No. 2, 1993a, pp.138~146., Design of Sequential Experiments for Contingent Valuation Studies, Journal of Environmental Economics and Management, No. 25, 1993b, S1~S11. McConnell, K. E., Models for Referendum Data: the Structure of Discrete Choice for Contingent Valuation, Journal of Environmental Economics and Management, No. 18, 1990, pp.19~34. McFadden, D., Contingent Valuation and Social Choice, American Journal of Agricultural Economics, No. 76, 1994. Minkin, S., Optimal Design for Binary Data, Journal of the American Statistical Association, No. 82, 1987, pp.1098~1103. Robbins, H. and S. Monro, A Stochastic Approximation Method, Annual Mathematics Statistics, No. 29, 1958, pp.373~405. Wu, C. F. J., Efficient Sequential Designs with Binary Data, Journal of American Statistical Association, No. 80, 1985, pp.974~984. Wu, C. F. J., Optimal Design for Percentile Estimation of a Quantal Response Curve, in Y. Dodge, V. V. Fedorov, and H. P. Wynn (eds.), Optimal Design and Analysis of Experiments, New York: Elsevier Science Publishers, 1988.