工作搜尋模型與失業期間─臺灣地區大專畢業生之經驗

全文

(1)~jf. .@.. ~. J:... 4'*$!f' 1c. I\t~~ $!f'1c. m-. 19 : 2 (1991) , 183-215. I{'Ftt ~m~~~~Mr8'. ~ ~m !fu~* _-_ ~ ~Z *~~. ** Iflil ~ *. ~. tI. ~ ~ ~~~- M~. ~ ~~ ~4. ~'~~ . 4 ~ ~~~ . ~~W ~ ~ ~~ ~ ~ 4:H'J~1ta~ . f. a. ;k;" t-lfJH.1. o-{'JJlJ. 72/ 73. ~a.. 74. ~a~. r i; ~Je. ~~l -f!-~t.t-t.. a~ ~.J a.4~~~~,~~ . ~tl~·#* · ~ . ~~ a. _~ ~ .fi ~ ~t~ M. :ill. :1f. BfJ ~.Ii. ~ 111! 0. ii tt .tJt. generalized gamma. ~. \'.9. ,. ft :It. -!f- 71" ~c. 'exponen tial , Weib ull, log- normal , ~ J:..* JJLii~ ;ft .prf Jlj. T. 1i;-ttf'J aiJ ~{~ ~.i~ JA ;f *-~. 1. a. 0. ~-~~ ~ ~*4 -f t~~~~~ ~ •• ~ .A tMM.~ ~ . ~'#~A~~~~ *M*.tl o M-~~,~ ~a ~~ ~~ T ~«.a ~ 1i; # . A A~ ~ *'~~~A. 4 • • ~ ~* .~~~~ 4 H A *~ ffi~~ &~o. * f'F ~ m iJ;t jiJ **~~ lJt~PJT Ii!IJ ~~ *Jl:t ~ ~. 0. 0. * XjJ; ~;f4 tt lVf~tttHI~~N SC80-0301- H004-03zM~ thh WJ ' • t\' ?~ 1'f: :g i7J~ij( l1i¥ IiJT :(ij' 'i'iJfjEft.J. 1'F :1§ jjz ~ ~#*iI1 /J \ Wft.J WJ ~I1'F :&~ lli: ti mAft.J 1U!. tij&l.o. .- 183 .

(2) 1.. •. is. §I Stig ler ( 1961, 1962) "& McCall ( 1965) mI±l Iff 1?1l~m ~ £nl~ , *=~~ &~ .~ ~I~ •• u~am~ ••••• ~.~~~ -o ~.~m~*= •• ~ E ~ tt~ ~.' ili ~ ~~ ••• ~ ~W'§I~~~ ff ~A~.Ao * ••• ~•• m A~, ~~. ~ ~ ~ ~ ft •• ~~~ ~ aM. o~ ~ *~~m* fi* ••• ~~ I fF f1lW~ f~ 1Jo 0.5HJT !Vf ~. * ~ R~ - .M.~I~. .... •• _mM I ~ ~ ~ ~ ~.~~ ~ I ~ ~fLl::. B~ ~fJi. •••~ ' *~~. ~.~*. 0. ~*=. ~~WX. '~0.nm ~~~ 0~~o - nOO~m 7 mD~*. W~~ ~ ~ ~~'~ - ~ ~~ ~0.m~. •• ~. •• A~ ~ m a~ -~. J!m~ nq~1ff~t~Ui* X ~ifjfN! ~~ I 3r ~ $&~. a ~ - .~~.¢~B n~. .... 0. , $~~~~~ 1~UE. •• '~ *~.R.M~••~n ~ ~~~, rr~ Ah.~. , ~ 1! 1Jo~:-~ OO~jlM. 0. I fF ~ iWm~ (job search model ) i.mt~~~ Stigler ( 1961 , 1962) ~~~.. McCall (1965) ~1E7G ~ ~ I{'F ~~tJ! ~ ~~ ( seq uen tia l search ) ~ ~£ ¢ ,. 0. 0. McCall ~R.i tf- @ j!U. 1t ~. it ~ ~ i:iJ £I. fth~, t±i -. @ Iii ~ e~. •. :iii ~ 1.1: 1* fllj. ( optimal s toppin g ru le ) : A5~m :;X1t.fi:l(* fO !i~ I {'F Z .~~. ,. ~:Q: -fW. l-*tWIR ( reservation wage) • R £ t~ ?£lj ~ fffiiT -1WI {'FI .~~ f* ifI. ' Ji';.t:Q:I!n*~. ,. ~fllHiHf4w tl(. 0. E! McCall ( 1965,1970) 0.1& '. ft~mbi@:~~ ±~ili1!. ~ -. 0. ~ ~Jl!iI~~~ m~ ~ Q.( ~ 11!I Lippma n and McCall. ( 1976a, 1976b ) & Mo tensen ( 1986 ). 1981 , 1983 ) £~ ~ tmZ;t:~, ~J:. ~*~ ~II $:~Z 1i~. ,. I {'F1t~m ~~~m. -. 0. ~Ei lm llliF;'z 7H. iJUlt~.i &**,ffiiE. - 184 . ( 1991 ). 0. 1*1J. ( 1988). m~-t. ~~ i?;t~~Z ]!! ~ I{'F Z ~:.

(3) .L it :ft4~t 1;1 ~!!-1;. -;t-JI!J r.'1 -~ i~. IijiJi. 0. J: jz(t)( ~!i\m ~I!flU!nIJ ~~ rm~~il. l~U~ ~~BZJ. '. Jt. ~ *- 1/- ~ -t- 1~~ .!\t'. {QW lliAZ ~~lj1\~CP ~± ~ft-J nt ~. (1'10 sLochastic process ). 0. 1*lil3.,J; ( 1988 ). ,. i!n~ ~ ~. ~JmI 1'F 11l w t~~n{1{. ~ g.~ ~. m~tt %E*w.~ J:'~~~. ~ - . ~ft-J . ~ o .~. m~I~ ~. ••~W~.MM~ ~~ .m. ~~. ( 1~2) .W~. m~*•••~ ~.M. F8~ W {*fJIJfz F8~ O'J Bmf*. 5im~~M8Jl ~. 0. ( 1991 ). ffl r~~ *H¥ zm ~{3 ~HIi!f ftl!. ~#.~~.MM~~' ~ . - ~ ~ft-J.~ o ~ I~• •• ~ft-J .$~Bcp,. • • ~ ft-J M~z - ~ ~~~ IJf fi~ . MM. ( unemploy men t du r atio n ) z~M{* ~ A'J Iii. '. 0. 'i*lj'Iit ~ ~~P5J 0- ~I f'F ~~:~Hi 1~Hj(. m ~ ~ [EJ ff.!f i~ 1JO tJUmIf'F 1l~ff.!fF8~. Oaxaca ( 1976) 8Z] ii~ PJ£Uj"A - Jt;i:ti ft-J~§~. 0. ,. i!n~ ~ MFa~. 0. Ehrenberg an d. {thir~~ ~~fl~ii'I~ fD ~~ AA. F89 az]fSl'7kff : (1)pJT fifJe~* ff.X * p~ fftzlZSl ~. ,. ~it ~{*ii'zI!t~. (2)ffif:f{i @ AI fm mtMj>O'J IZSI ~. ( 3)pJr~{Jg1lmAIfF13t iJijt~~-~. ·. ffi)~{Fel*ii'Iit~ {~. , WtW{~1*1j I ~fl1 1JO '. ,. ~JJX~ .M FI3'f,fEJ~ ;. mtt$fC.W1 FI3' z~ U;iqrUE. (4)pJr ~.fie~ttI {'f.11f pJTrm~ft-J I if?tWC IIiilt~~~ ft 11f ~ M rEl9. , £l ~ ~Ur8' !I1 1Jo ;. '. ;. ~~~Djlj ~ii I jf fD~. 0. Kasp er ( 1967) 0. ~~jf*,Hi5rr jJiJt~:P~{,[ ~ /N:;'f I jt O/~. ( 1975) ~IH~~lj 0 .2%. 0.375%. '. {th~5f!:Jt m ~ ~ AAF8~Hi ij [email protected]. '. ~~~. Toikka ( 1972 ) l~ ~IJUf.J{*~ 0.1 % ' Barnes. 0. Crosslin and Stevens ( 1977) .l'!1j Wc!.~ {* iiI i'f~*~·. 0. ~.mM '® ~.AAM ~AmX*.~.~ ft-J ~.I • • ~ .~I~~~ li~~ ~. 0. =E0-:B.-/J\ Lji1J$ ( OLS ). {3~i·. ·. it il€~{fH5 ft-Jf.i!i ~. 0. (th{f!~tl%~~I ~ fD ~ ~ Wl. M.~.~.' 0- .~ 1J ~ ~ *~~ 'M • • • ~. M Mm. ~-. f.l' § ~I . c,ur l~. O.3%flj 0 . 6 % Z ~8'. ~ iiifl~. ( 1992 ). 0. m ili!t ttl!~ ~J:~!l/t~zjt*4 {ti~t ~ m*.-$~1:JJJ ArPJ~~. ~~-wJ lb'4ij±@1Jll -@I f.l. '. ,. ZP±S]f*-mIifi" r ~~ O .32% ~IJ 0.7% z FI3~ , ~~~ Z. - 185 .

(4) ~ ~ ~f®j. 0. -1JooPJ n~:Ji! * 1#H'!~.1: mAm ~. I.'~-1Joo~PJ~ . ~. . ~~* . m~~~ A ft~~~,tx~ Mw I~~ ~. f.£rpJt~ ~~ {~1il ji fi~ ~ ~ ~ jj. El 1980 ifftD.* .. . tx~}~C1' R . ~~~fi~ml~ f~1i. mt8littiofilT '. 0. ~~m~f~HiiA~~-~zliJf9t. 0. ~ ~,tjH~:,*" D. Il! :t1.\b£'lf.£t~~*ta'ii!:tk~~AAFa9 W f~1¥il ji Z. Fs9 Sl ~m {* • ~o Kiefer and Neumann ( 1979. 1981) . Lancaster ( 1979. 1985. 1990 ) . Sharma ( 1985) . Kiefer ( 1988). 0. ~*L. '. fili1r~~.~*=~M Fs9*Jit,§,. fi~ ¥ -.*~o~*= . MM~~~§~W~~I .~ ~~fi ~ ~ o ~-1Joo,~. ~ .m 0~~*~~1J~*~~ *=. MMo~.tt~I~.~ft~ . MM~~~. ~M. '. ~IJm* ~ M Fs9~ ::f ~ fi ~c ~ t! ~. exponential *~. ( memory less ) • llt ~~ ~ AA Fs9 11 ~. ( exponential distribution). 0. ~t!<J~IJI fF~.~W*~Wl Fs9. fi roM • Nil m~ ffiI. ~ Il'if Fa' It< *N t! (time dependence) • Weibull 7t ~ ( Weibull distribution ) ~~~~~!:t~1r.. 0. $~ L. 0. i§:. Mi ~. * M Fs9 II m. ~i§:.iI.ft.1~~I ii 2J} tt l~ftHJ*. WIFs9. ' exponential. 7t~ft Weibull *1'iCm~ generalized. gamma 7i'ifr: ( generalized gamma distribution, GGD) ~~9-*~@. fi f~ '?ti ,lJ:. *][ m Jt. ~ 110 £1.. ft ~t. '. 0. ~O.tt$llI{1:~.~fljTI. .§.rJt.~1r.~* E3 {~1¥il jt ~~ 1r.. uHJHt ( true time dependence ) ;. ~ rJt fj~ 1r.1**. .. Jlrjm~~ ¥n!r lF.1rI39. El ~fili~MiFa9fi ~m ~ lZSI*z. ~{I::. ' ~~.g: A fI ~¥rjtt¥rlI{1:Z.$jljfj*~MFs9~~1r.. t! ( spurious time dependence). IZSIllt-ili. Heckman ( 1978 ). Heckman and Borjas. 0. ( 1980) J:{rjJt~ ~ Wl' Fs9' 5t1'iCpJT ~ ~~MiFs9*ij!t!1JoD.~;'jlj· ~~MrI39 ~ 1Jo~ g[ !:t~ {1::.. 0. '. QrJm~~~t!£~Fs9{t(ij!. 0. *liJf9t~{t(_M~I~MwW.,~ ~§.~~* • • • !:tmA~.m~z~ tt'~R-OOfi ~ ~~~~IfF~~mMo~ ~ ~ ~ z~~~rr · mM'0 . ~1J ~~fi* •• ~ !:t.ttlfF~ M tt o ~~~~~fl=@~a1Jo0.M: (1)~. 1*$ • •. (2)*• ••. !:t~ .MrI39~ ~ ~±: ~ [Zg*~filT ?. !:ttt$rjI{'FZ.$ , ~ ~ ir jljfj~.MFs9ij1 ~ ~itJ1109X ~j, ( !!n~~filF.1 - 186 .

(5) ;L. rB~{t~ *JH~). 1t :It 1,f {~ 1'! ~ 9<. -t ,I\!J r·.1- {. ~. il:. @,. A __ -f it 1.. -z .~l ~. ?. (3)*__ "' ~~~t~~IJ~~ W~~Mra'~bt~BJlM~~ ? *~~.~~:m=~B~-@a~ ~ . ~~~I~M~.~·ft~~~fi*. • • • ~I~~~~~~~a~~M~Htt~00fioX.~M~ • • :!It~~. 0. ~.a~~. m-=.mJ1*ffl r:p1 ¥*,JUftflilf9E~72 / 73:&74if~$.L-!~Hj(~~~ :N · {6Hw3. l};;:~MFB'~IZSl*. 0. :ft1r~#f7t55ljft. exponential, Weibull, log-normal,. generalized gamma ~0fR055IJ1JD0{r5~t. .. :&. 1fZ.l:ttt~~~ 0~~~Fa%UjH~rt:Jff. ~ . ~o~~¢n~~*~rt:J-#AA~1fZmlli~0~- ~Iilf~rt:J~~o. 2.. If'Ftl~~~~. 2.1. ~iEl*-mIjtT~If'J:tl~m~ W~*.w: .*~~ fJ(:If'fJJt*m-=-:X-@Iff::S<]~i\:. ( sequential. search). 0. ~titt~-@Iff::'~~~£~m~o~.m~'~~w~~.~~~;~~m ~,~ .~ tt a ~-@Iff::. ~*. • •. o~~~£~m~~-Iff::'~~~. •• ~fiofi. • •• ~~tt~-@Iff::,~ •• ~~,~*tt~~Iff::0M'*. "' .~. /G j;D ~;1iI; ~ iE ~ {@: (realized val.ue). ~~. Y , ill W~ !=iff ~ ~ 181 ~.i¥ 0 fR f. ( probability density function, PDF) , :&~~tl~.i¥jf;I3tITEittF ( cumulative density function , CDF) f'Fz~~I1X*. '. 0. Mf~J::.~~::h. ¥Jfi~)t4.ijfJ(:~IJ-@If'Frt:JI1X*~ co. '. t-D~~It!j:rt:Jifd: TIr :&JL,W ~ ::h. 0. c EH3~ J\t filiI. tloJlt '. ~~ w tlt n. @I f'f 1~ S'll'l M¥IUHll Y~ ~ : Y = max (WI,"',\hl. n ). (1). - ne. ~fimR~*."'.~~ r. ~ w ttIff::~~ J ~~~.~rt:JM ~ ~~. (expected return). 0. jffifi~ WI ~ ~ tt3iIJm-@I ff:: pfff~~I.. - 187. ' Jl'J M?!\ R WI.

(6) ~r-@JIlta/JfflMW~JH. rule). 0. ( optimal stopping. J¥;tPOrpjfjz!); :. m~. w,. fJllH~ "1¥tx. 'PO~. W, ;;;; R. ' PO :W:. W, < R. ffij{:&ltt &tH~lt ~fIlj. I. IZSIffij-@~rm ~B1wttIft*~ltaJliJ'. ,. (2). :t£~-:;j:tX¥IJI fFZfljt~ WI. Zr '. lIJ:ti:~tt~-@). f'Fz 'f:ffM fR ~H ~ E max ( R, Wi ) -. (3). C. R = E max ( R, Wi) - c ff(4)J:\;*~~'. 14f¥IJzR. (4). E!n~pjf5\'jB1{*~Iii{ (reservation wage) .. J'JIIX:*. c. W Iit vV7-}~cOl~~ (ttl) •. (5). R = R(c;f). lttdR~fimwttIftB1~~~*cfiI.B1.*~~fft~~~.MM§ ~B1~ ~ '~R~~~*.MMB1~ . ,~-~.oaffOO-~U~'Iftg . ~ tj(:¥IJlf'Fz~~~ A ' ff!}~m~*. ( hazard. rate) (li2) '!!n. oo. A= fR. (6). dF ( W) = 1- F ( R). S~& I.a ~ ,~~~~~,~a~g~tt~I~,~ltt~.MMa&o~. ~*AW* . MMt~~~B1~8H~,ffij*.~~.MMB1§~~a~-.*. ~~'~.*m.~.~g(t) 'M*m~*z~. 'a~*.B1.#r:t£§M. - 188.

(7) .:r..fF 4t4#t~~!k -tM rJl-.t·~· J~ ~ * ~.f-t 1:..t~ .~. ~~~~tt~I~~.* m~.$'~~'~ ~ .~~~:. g ( t). (7). A (t)= I-Get) ~WZt. ,1=. get) I-G(t). ~:¥;(6)it~(8)it-§-{*. rt. Jo A (. HI. = A • ~(7)~)ZPJ~ffX. A ( t) W~Ff3~mt/U] , fI/1 A ( t). A (u). '. u) du =. A '. =. (8). RJJPJ1~fIJ. rt. G ( t ) ~D. . HP. get). Jo 1 _ G (. =. t) du = -In [ 1- G ( t) ]. ~. fl::AJ:.rt '. - 1n [ 1 - G ( t) ]. g ( t ) ~ilEi ~ Jf5J:\. fo. t. A du = A t. ~. G ( t). =. 1- exp ( - At). (9). g ( t ). =. A exp ( - At ). aO). E8(9)Af~w1iJi2~~.$ A ~-11\ ~. ,. JtIJ*= ~ WlFf3~~1f5 IffX-@exponential5t~c. '. ii~B~*=~WlFf3'ZWl~~~ 1 E(t)= ,1-. an. 2. 2 . l!! lI'ilra'iOO ~ f*m I jfT£r..JI 1'Ftt .m ~ lIt~ =¥.Itfr~W~I{'F 1l~ ffX*£~rf3~Ef;J~lt. Y = max ( W,,'" , W n. ). -. n•c( t ). - 189 . ' c=c (. t). ,alJ ( 1). ~PJi&ml1X. 02).

(8) (13). R = R [ c ( t) ; f ]. (6)i\ rpij';]. it ?)\ ~~ lMffl'l' ij';]~ f{ · it ( t). fR. 0. co. it ( t ) =. 1It1Mf G (. (14). t ) 1J.X;~. tJ. G ( t) = l - exp [ - il ( t) •. C( t). = 1-. exp [ - ( it t ). (15). a ]. g- ( t ) = a il ( At) a-'exp [ - ( At ) .1t~(9). ;fJ }. dF ( w) =1 - F { R [ c ( t ). . ([5)PJm . Weibull. (16). a ]. 5HR.TJ~ exponential 7Hicij';]-M~1~. · PJ5tIJx =fl tllf. iR: ( i) a = 1 ' till Weibuli. 5tNCmftiV<; exponential. ( ii) a > l ' JtIJ ~ ~ Ili§5'C~ ~ r8~ t. !j1Jo ' G ( t). 5t ~. ~/J\. ;. , ~~MPs~ ~:&: a':j;t!$~/J\ ,. Em ~ fiNJaij IfF ; ( iii) a < l ' ~jf:; IIJti5'C~ lMfrB~ t :@ :;r; ~~~IJI ff. ilj1Jo ' G ( t). ~*. '. ~ ~ :ltFWB~ ~~ ij';]~$~*. 0. fi~ PJ0 H~ tiaZ~~'m~5'C .AA Mti~~I~.$Z ~ ti *A~.MM.~ 7t~Z~ ft(~li fi. ( t ime dependence). $I.'x. '. '~ ~. MM. '?)\~PJm~* • • • ~ij';]~ . AAM~~A~~M 0. We ibull 5t~{* generalized gamma. 5HJc ( GGD. ) 8q~7-*~1L~. 0. ~. ffi *~. m~~~ *• • • ~~. MM~ft~~~M~Mtt.,*~~. ~~~~. - 190 .

(9) .:LVF it4-~~.4I!-1c..1"AI] M-"·?t Jc. ~ * ...... t - ±z ~*. -~il:i t ~ 7t ~ ~ n!%. 0. 1 ,.., (e). g(t). Et-J ~-%il!n!t i?illfA g ( t ) PJ~ rtt. ( GGD ). a A. ( At ). 0 a-I. exp [ - ( ), t ). ;!'t cp GGD cpfrwHI~±~.t5fi : a ' ~mt~ ~lt. JtIt ~ ~ (s hape parameter) WeibulI5t!!2 ;. Ii () = 11i a. ~~~ iliM 1 '. = l~. JtlJf5ilt GGD. ti ~ ~ ?JF {@: f~ -. t~. 0. 0. tE () =. ~. (). =. =. parameter ) ,. e'. mJf~. llJiij: , GGD il;t :J8 it ( d egenera te ) 1Vt. , GGD il;t il8 itJJX exponential 5tWC. ~~ ~IJ Z Fo~ W1 J.t Wei bull. B-. 0. :lQ f~. aIF!¥ ' :;Ef!t a ~ 7]0 if foHll ilitJ '. a A. t -lexp [ - ( A. t ). aGGD 5t~ r:p·. 07). ]. ( scale. Log- N ormal 7tWt: ' ;!t~i¥ {t; ltl®Jlt g ( t ) /.!X:~ g( t). a. ( ;tt3 ). ~IJ. 0. M?,).\ :;Ef (). , GGD. Er-J~. GGD il8 1t 1Vt. : (18). a ]. •••• aft~~• •. () ~~&.ili 5t ~.~a~ ~ M ~. ~ ft 'AA0 ~ft ~ fi R .~~ ft Er-J o ~fi ~a~ () Z . ~ 0&~ ~'~%~. Weibull , log -n ormal,. :& exponential 5t WC ' 1ZSI .Ii:t/.fs:X::W,t5t7:l IJfi exp onential,. Weibull, log- no rmal. ' "& GGD 1]O£Ut ~t. '. illz*i ~:;r;; ~5t1fr:T jl* ~~ f~. ( likelihood value) Z* 'J\ **Ij JE(iiJ ;fI7Hir:jl #lHilth:!1;* $!f!*~ ~ ~ JtEl rl3~ pJT Jl:f:f zM: ~o. 2 _ 3 . ~t jHl ~. tEAA~MM~ ~ r M M J ~ • • 7tfi$,. ( censoring problem). 0. • • ~~ft. ~ .~~ Er-J M a. EI3 M~ ~Jj flI3 ~ Mf Fe' ft:J ~!HliIJ. ( pa nel data) 'lli .tz;;l~tru~ ll=!frEJ' B'3 ~*. ' ~ rl'l9 fI} ft 8":1 *tHl rID ft *4. • ~ ~~~ ~liM. M rl3' 8'i =1' 5C~. o T t;J ~ BjH~\i! . fl{f Fe' 1ft!m IJJT Er-J FQ,~ :. - 19 1 . 0. PI.

(10) t. to. 1. ft ~ fiil. I 1. To. : to. t1. .. 2. lN~ M2. :. 2. • • to. t1. 3. 3. .~ ~3:. to. t1. .• .• ... 4. D~~4:. 4. /lj] 1. 1Il1r:jJ, to, {litm i @ D~ ~~~~ 1Bl&~ zlMf~ ( :tm~ "' IMf ~B' ) ,. {tlMfM To 0. !.%. £nlr~lI'!ffk~. 0. ~tE Mr~ j()J~IjEtJU~IJI{'F. tIl_tO, ' ~ -7'C.il~Mrl3'. ( co mp lete spell). 0. t',. ~~tJU?IJI. ' ~p t', ~ T. ' ~1J;!t~~Utfj rl3~. ~ tEfHrf.*1Mf. '. f6)tE~ ~ Jtl\:~. ,. Wt' ~ ~,~ ft~R~ m .X ~tE~ ~'ili .~.mXW_~~tt~I~'~ f!!~AA rl3~ ~5C~. rl3~. ,. ffiJ B~ M. ( incomp lete spell). 0. tml;! 1 r:jJ ,. 2 ' 4 Mrl3' ~ ~ 7'G3t inJi Mrl3~ ,. censored) (H4). ¥i ~ ~. 1 ' 3 ~7'G~l¥Jfi~M. ~1f~m~=Mrl3~~;:fJ1J WJItiJf. ( right. 0. ~1'r~ ~~tmt~ _. FB' To Jl!!fl1kWJ rl3~m 11. 1i:IT: ' [ZgffiJ :ftfP~ R ~ -;,l5 Lt: fi ~ M FI3~ 1¥J;t1,t. $~~'ffiJ~ . ~ ~ tm . _M~a~~;t1,t$~~(tt5)oti~-~.MMt£7'G ~ I¥J. '. j:j IJJtit3t §B ;jftfj;( illTIlt l¥J ~,*!'%~~ WJj[illTI~. ( PDF). 'g ( t) ; ~~W[ ~. ~7'G ~ ' ~ ~A.~~~~;t1,t.~MillTI. '@W~ 0 ~~~MMI¥J~~.$@ ~ illTI. It ( CDF ) , G ( L). 0. [Zg~. , ;jftflfJ. i?iifj: r.iJ ~ gX. - 192 .

(11) .:L1trt. 4- ,j~~. ~. L = I1. J.I!-;l( i:M r.,-. {· ~ ~ ~*. ....... 't. ~.z ~~. [ g ( t;) ] a, [ G ( to) ] 1- a,. (9). i- l. (9)i\;cjJ , 5E'. to ~jC~WJ Fa'. i\;m ~If£ § ?!.\f.t ~. , ~iJ. a, = 1 ;. '5. to ~ :::f'3¥:~ iH~ AAFa'. , aiJ. a,=O. ~(19). 0. , f~. ~. 1n L = L { a , 10 [ g ( to) ]. + ( 1-. a, ) 10 [ G ( to) ] }. (?O). i'- I. W#LI: ~IHi 1$ MFa' I¥J ~ ~ 71 fI: ' 1ix A (?O)~ cjJ , :£It 1r~ {j[iiJ QL fl Z 11Il QL {i5 ~t. ~ $:. 0. (20)i\; cjJ ~ ~ItI¥J!i *:ff!H£J. {t~t ( ma ximum li kel ihood estima tion. MLE) ' :ft 1r~tE(2o)i\; f.t~ ~ It {~~- ;;R fo~-=-* ~ 71. *m. 0. ~cjJX7t£ta 'J\zp1J$. ' Wue ffl Newton-Raphson. 3j(~~i~]{1i. ( OLS). (initial value). SAS/ statistics Z~~~.J\ Ml t~i\;r:p tJ)~ mfi.X J'J'~ {t~ f¥APJ #t~ffl. 0. &:mt!*. ~J!I;:1J $tE. 0. Jl*4~BJJW;m$-jl~*s*. 3. 3 .1. ji *4 *~W~tiJEfi. *~~ue ffl z~~* § cjJ .~~~~~~~l¥J r $ L.~~~~ ~m~ ~ J o ~• • ~~f.t ~.~~ . ft0~Z . ~• • ~ • • ( .~& ) .z~.M~. ~8 mJ ~. 0. ;!t~ rrm ::!RmJ ~. 0. "'• .( ~~& ) Z.w:~ ~~ !:t11D0 ~1f =::R1f: ~~ 75. '¥ j(;Fl ~~ Fc~ ~ iJ-ti. ~- ~ 1f: ~ ~ 74. 4j(;ftJ ~~Fo~8 ' tt-tt 4=" = ... [9ftj. 0. FQ~~ 715j IJ 1n 74. 74. @] J&. t. 4 = .. .=:.. 4 " . (~H~&). 4&. .. !?3ftj. 0. Fc'{(il?j-5jIJM 75. 15. 0. ~cjJtEmJ 1f·~ ~ IE ;ff;tdl cjJ ff.] ¥3 ~ ~ 1300. 73. @]J&. 0. '¥ ~. Z "J:~. !:t11DJ2),~. 4i~~i!fZfB'& • •f@:¥3 14000. •. 0. 72. Ah::ti ( H6). @h::. 0. tEf.t*~~~mz ~~ ~0~-~m~ z~· ft~ ~~fi*~H.I¥J~~~ ~. §'~ .MM'~QL ~ ~otE §M~ •. •. ~WI¥J B.R .~ •• z cjJ' ff *~~~. . MFa' Z {trr ~ 71ffT iEitHr l¥J ~m jdm~ :{:E1!7St I¥J ~ iIJ~~. ~ ~'fJ {JJ?J\0 ii1f~ ~iIJ 0. - 193 .

(12) * ill ~. ~~U.a~'fi~.W~~fi.P o ~ ~~~~-U.. M~~ - I.~~ 0~. •• ~ •• ~m ~. J •• 7 - ~m~~~. o. a7m~~.AAMA~. W&I •• ~~~~,*~~m~ fi~.AAM ~~~. ~ .W.M o ~ r .~. . o ~~~M.¢,&~~ r ~~~~Mtt~.-OO I ~ J , ~ - ~~m~~ A . .~~~~.MM o ~ ~ ~ft -~. ••~§ • • dm~.~*~ ~ffM I ~ 'fi~. ~~~*'*X~~~~~~M~~~~~M ( d~$~M ) ~~~.MM'@~ ~¥IJ gglWr~.*4~H1 z. a. mtEftfr~re*1iJf~pJT~ffltr.:Jtt!~:gm&JE~7t 55 IJ1J!]Q{~BJl-ttDr. AF : ~19:. a. * AfH§t ~::f£'A R1§t~. :. , AF =O ; ~~~ , AF=1 a. AGE:$~ o Q{~~~M~~ ili~ ~M nrr ~ zo. BDEN :fi ~.~~. ~a~ o ~~m. •• ~2~m.Eft~ m ~~.~ m .ffl'. Qlj BDEN = 1 ; 1§QIJ 'BDE = 0 a DUR: • • ~~ ffi . tt ~~ - @ I~ ~ ~ EDU : ~W*~. 0. EDU = O ,. n !lH~; EDU = l. EGMY : ~~~1f;:ill'f.Ji& {lf If'F~~ FIELD: Mlf; MA. 0. ~ _M(~. )o. ;fi~~ 1: af!'. : ~! tlW o ~*~~~~. iii I. '*~!ilG1iJf~,6JT $.. 0. a. ' RIJ FI LD= 1 ; fi RIl FIELD = 0 a. , MA = 1 ; E. ~ 'AlE~. ~D!f~ ' MA = 0. a. ~T H :* ~~ ~ o~.~afi ~ ~A.~~,. ~ afi tt . g .~ .ti ~~ .. ft ' IlIJ METH = l ; fil'IIJ METH=O. 0. SEX : -t!f,lj a J.JM: ' SEX = 1 ; ~M: ' SEX = 0 a ~~iX (f1zp.±S] fifIJrJ 1i{-~. 1a. ~~Wlrs'tr.J01E ~J ~ 1II. 2 .111 3. ~.l ~ . ~. ~ ~ ttl'f.J~OO.~' M. ~ fl .~·. ••M•• ~~ .a. ,*. ~.Q{~~fl~ ~. oUfi. ••. a. ~ WI ' ~ . '. M. ft~ nrr~ ' ~~~~.. 1i{-~ -t! ' @$~ ~ 10 %o~ . ~ ~.~K& ~ ffl~~~.~ m ~ .M ~ffi' n/. 73 $ ~ 13% ' 74 ~R~ 10% ' MjjU!J L .~~~*~IX {JJ ?i\ {~~Jta:Fs~jt~~ tt. - 194 .

(13) :L.11 :li....q:;j*~ *- ~:1.·J\fl rJj -. ~1. ...· ~. Jt. li Lk4'" -t i.-t~~. 4 ~ ~g * ~. t~~~~ ~*. 74 ff. 72 /7 3 ff~. 4~ (AGE). \lt. ~tt"R. tt~. a~. tt "R. t!E. 26.580 27.95 6 25 . 138 25 .403 26.696 24.0 16. tt~lj. (SEX). 0.512. 1. 000. 0.000. 0.5 18. 1.000. 0.000. ~t4J;l. (MA). 0. 83 2. 0.75 0. 0.91 9. 0.8 59. 0 .790. 0.934. J'i:i5t. (AF). 0.442. 0 .863. 0. 000. 0.48 0. 0.927. 0.000. H* (FI ELD). 0.643. 0.7 16. 0.567. 0.66 0. 0. 744. 0.571. ~ :;iiHi ll. 0.530. 0. 562. 0.496. 0.5 76. 0.618. 0.531. *~ 1J ~ (METH ). 0. 129. 0.137. 0.1 2 1. 0. 099. 0.098. 0 .1 00. ft l!UJHHf.f (BDEN). 0.060. 0.085. 0.034. 0.045. 0.059. 0.030. 4J ~~AA ~ I Jt (EGMY). 12943. 14956. 108 32. 1377 0. ] 5903. 11481. 1.793. 1.888. 1.69 3. 2.726. 2.430. 3.043. (CENSOR) 0.044. 0.042. 0.045. 0.05 9. 0.060. 0 .058. 11175. 572 0. 5455. 9006. 466 1. 43 45. (EDU). :k ~RJtllr8' (DUR) (a ) iiB~ 1W:l&J~!l1filt{7lj i!B~~f.!H§. ~ : ( a ) ~~m£§~~~ * . ~~ * .MM · ~~ ~~_ E~ m& o. - 195 .

(14) %. o. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10 11 12 13 14 15 16 17 18 19 20 21 22 2 4. ~~H;Utl1rB9. &i] 2. 72 / 73. ( fI ). -+ :k t- -f it ± *'- it 1111 (q' -=Z. ~ ~c.. %. o. 1. 2. J. 4. 5. 6. 7. 8. 9. 10 11 12 13 14 15 16 17 18 19 2021 22 23 24. ~~WlrB9. ~3. 74. -+ r.. . ~ -ff ± *'- it 1111 rq, -=Z. ~ ~c. - 196 -. ( Jl ).

(15) t· ~ jl!.. ~ ;idl·~-t.:i. z#dl.JJ&. .:1- 1'1' tt 4~-iMl ~ j( -!- JlJJ M-. I 11:. 0. 3. 2 . ~ ;~um. rs'Zfi!irr. ¥::$:J: ~ @l~!4.rAA r8~Z I3l ~ri5J 51 fi..t=* ~. : - JJh~: *.$;$~~:;$::!it pJi $1 ~ Di !f,f. ti' tzoifOO" ti5]IJ " ~ ~. ., HtJjlj " ~(jf;H~JJt" M * ~~. ffruj81 tJS1~. " ltHUU~l.m" &;\J~lF.fzJtl'1~Iif. , :tm * I!flE :1J~. ; ~ - ~~*§~I {'F jf~ M~tl~JEI ffl~~. c. ~OOfi ~ffl~ I .~A~ ~~ft*EQ~~*.~~~~0~ ~ lii7tru:~~ lt 1J.ft!£ ;e;. 0. ••~~ti ~ m ~. ~I {'Fitw~rm~I'fJIjfftWCfJi~0Ut. i~Ed~J.ttgt\t~.jIJ:ljJ~UIf'Faq~1¥. '. i:Q,~~jIJfjU~lJIf'FFfJn~r8~. 0. '. t§l~I f'F 81/:i. £*J:.ffri §. ,. 00. ~~ -~ m . I.7t~_ ,~m FfJ I.a.'U.~3Qtt~I W o .~~m~E I.~~~:Mr8~JJm~~ .~ iI l± tJ':J. 0. ~71-ilt1P'1ffmj;j]~.~~FfJ*=.M Fci'J~"l\iitB I~Ib'ft(iHi (time dependence). ~ ~£*. MM~ A~'.~.a~~I~ ?. ~~~Q ? UMM FfJ W. · ~~~. jJ,flj ffl~[5]lY-Jm~ 7t IfC*MJE[tij:r~'{i(tIH!Ul t!'m:. Weibull , log-n ormal, ~J,* ~~Z* !J\. 0. 0. 1ZSI.llt~J1l~. * J1IJ ~. exponentiaJ.. "& generalized g amma 1Jn.l2),{5tl- • ~~~:jUflt~~~W. , 1tIiJlHfa1tJ.lif! ( likelihood value). *~j;EfoJfit71l'£ :Mdi~tt~*4. ~0 m ~ o ~ ~, ~~~~ •• ~ m ~¢,~ttw~ttlY-JI.~~.* FfJ ~.'~ =tItir~WJ)~filifr5 tEI 1t ~~ J: -tB itffpJT::fj6]. , PJT .l2J.~Jf.t. 72/ 73. Ez. 74 i¥~mMl~. !£ .~ ~.l2J.n, ft ~ .~ 0 ~ttn~tt ••~ ~0~ff·.l2J..~~~~M~~~. ~!'f.] ~~ ft~. illz~M%. 0. i@$ {t!lIT~*7IJtE~ 2 ~~ 7. 0. 2" ~ 3 WJ~JHJQlfUi!HI~¢ , ~fj¥~~~ ( AGE ) ti~~~Wi rp~~~. '. im?l\ {-th{p~ 8~rm%~lHiH~lA\J. 0. ±~{f:JA':tlt ir~Lf..Jl1tBi~~~~~~Ij'I~ftOOt;{f~. ,. ~~~ .~.~~*'~Xiffifi~I ~ •• &~.M M~~.~Q8.~ * o ~ U~W~. ( M.I\ ) flnt~~jtijrJ1~~.i¥';:BJHm. *~AArdl~~ '. 74 ~B\J ~*4tP. '. 0. frNij*Jl~'MtP. ~i*IX~F-m-g~Et-J* l~. - 19 7 . '. 0. *ug;~ 0. (MA=l ) Al. ~~ 7F tffi~~~~.

(16) generalized. exponential. Weibull. log-normal. ~Jtlj(CONSTANT). 1.87030 (13.641)**(a). 7.02614 (10.947)**. 11.35781. ( 12.556)**. 1FOO. (AGE). - 0.04054 (7 .896)**. -0.301 74 {1 2.740)**. -0.5 7124. ( 17.235)**. ". (MA). 0.25632 (8.588)**. 1.071 61 (7. 526)**. 1.72657. (8 .73 1)". 0.05 423 (1.102). ~m: (AF). 0.04582 ( 1. 145). 0.46881 (2.5 15)**. 1. 19 192 (4.670)**. 0.06208 (0.871). tt~IJ. (SEX). 0 .39675 (9 .176)**. 1.61 770 (8 .091)**. 2.94288. (1 0. 795)**. 0.342 11 (4.493)**. ~Wf¥lfl (EDU). 0.11 06. (5 .229)**. 0.72468 (6.773)**. 1.18670 (8.032)**. -0.00982 (0.267). - 0 .10119 (4.864)**. - 0.22012 (2.23 4)"'*. -0.30832 (2 .277 )* *. -0.12756 (3.5 77 )**. - 0 .01377 (0.333). -0.05641 (0.289). - 0. lO935 (0.409). -0.00622 (0.085). - 0.04257 (0.307). -0.33879 (1. 773)*. 0.14607 (3.060)**. -0.00002 (14.700)**. - 0.00032 (17.728)"". - 0.00003 (7.488)**. *4*. (FIELD). ~.§!:tt1t (BDEN). 3J(:Jm:1Jli (METH) ~~ I fif(EGMY). 0.19911. (6.819)**. - 0 .00005 (22 .802)**. (Scale Parameter). 1.00000. e (Shape Parameter). 1.00000. -63060. 15. Q. Log likelihood. 1l. 4.72142 (92. 980)**(b). 11 J 75. it :. (a) t.5~ Pi ~ t fiUiUt~. ( b) t. 1.721 35 (7. 25)** - 0. 00675 (0. 739). 6.555 49 (1 24.22)**( b). 1.59329 (22. 69J)**(b ). 1.00000. 0.00000. 4.04931 (4 1.613)**(b). - 344 3.98. - 35488.05. 111 75. , ff * '" &. * :1l :iFjfBII~.IE 95%J!ij. fiI'I!Jt~mf*-'Ullf(m'U~ 1. gamma. 0. - 198. -3 J903 .52. 111 75 9O%ZgW*~. 11175 0.

(17) .I.{t ~-4-~~ $!. j( 1." M. ~3 ~~"J!~. ;k. -t $.J) rEi' -t ~ if 1'5 it. :. rolJ-{' it J-I!.. g :k...... -t ±-'t..~ $-. ~ it .k 11- -'- -t:i. 74. Jf. : 9:2 ~WJIb' ( DUR). ~$ )t!iC. generalize d. exponential. Weibull. -m-ItJj (CONST ANT). 1.59833 (1 0.197 )**(a). 5.96079 (7.793)**. 10. 12903 (9.221 )* *. iFti. (AGE). - 0.01589 (2.669)**. - 0.25347 (8.582)**. - 0.51 265 ( ) 1.880)**. ~Mt§. (MA). 0.4647 9 (1 2.920)**. 1.1 6002 (6.860)**. 1.69947 (7.167 )* *. ~~ (AF). - 0.1 21J2 (2.067)**. 0.62580 (2 .280)**. 1.7029 (4.457)**. - 0. 25 108 (1.753)*. fj:}JIJ (SEX). 0. 17 538 (2.9 J 1)**. 0.84919 (2.982)* *. 1.7568 7 (4.417 )". 0. 18445 ( 1.259). ~~fjJtr (EDU). 0.03934 (1 .488). 0.47262 (3 .7 13)**. 0.955 77 (5 .294)**. -0.04182 (0.670). M~~m:. f F(;. log-n o rmal. ga mm a. 2.0 ] 57 .. (5.65 1)** - 0.00880 (0.657) 0.48886. (5 .860)**. - 0.05727 (2 .417)**. -0.1 6256 (1.447). - 0.1 1557 (0.735). - 0.09537 (1.684 )*. ~ 1l!.~jt(BDEN). 0.2 151 0 (4.042)**. -0.33 675 (1.353). 0.63740 (1.83 6)*. 02981 (2.325)*. 3J<1!a/j$ (METH). - 0.0991 4 (2 .762)**. - 0 .56100 (3.320)**. - 0.93 744 (3.902)**. - 0. 29453 (3.709 )**. ;fff11lI ii (EGMY). - 0.00004 (3 1.135)**. - 0.00014 (16.l01 )**. -0.00029 (1 7.318)**. - 0.00004 (1 4.624)* *. (FIELD). .. ex (Scale Parameter). 1.00000. 8 (Shape Paramete r). 1.00000. og Likelihood ~. -50628.95. ~~1~1t §. 9006. tl: ( a) t.5 WP'l ~ t fiU l!tHl • 1'f * * &. ( b) t 1ia!!Ji ~ ~f*-Ii!!lft l:t-J ~ m- 1 0. 4. 693 55 (81 .1 36)**(b). 6.7 1000 (109.79)**(b). 2.Jl070 (30.995)**(b ). 1.00000. 0.00000. 3.02311 (34.872)**(b). - 27837. 12. -28683 .77. 9006. * :~\hj~7tjj IJ ~.If.. - 199. - 26498 .63. 9006 95% ~ 90 %Z. !!l~**. 9006 0.

(18) #..iJl-l. *.4. !k 1;" jJl raj z~!f#~t. :fNJW,,~1t. Jj 11 kl} -l- 1;" ±. 72/ 73 -f. : ~~Mr~' ( DUR). #i$5}~. exponential. Pt¥*, ~ f,{. Weibull. log-normal. generalized gamma 2.21610 (6.247)**. -m-~IJi (CONSTANT). 2.43886 (12.777)**(a). 9.19957 (10.734)**. 15.39418 (12.844)**. 4~(AGE). -0.04181 (6.451)**. - 0.30160 (10.540) **. -0.58667 (1 4.659)**. ~t@(MA). 0.27467 (7.537)**. 0.96461 (5 .842)**. 1.58 166 (6.939)**. 0.19703 (2.884)**. ~{)t. (AF). 0.03577 (0.887). 0.40292 (2.246)**. l.l1521 (4.575)**. 0.02903 (0.378). ~rn~oc (EDU). 0.04225 (1.385). 0.40890 (2.897)**. 0.50351 (2.586)**. - 0.08949 (1.584). - 0.19987 (6 .545)**. - 0.29682 (2 .158)**. - 0.20195 (l.075). - 0.22619 (3 .855)**. - 0.05481 ( 1.11 6). - 0 .33885 (1.518). - 0.76357 (2.530)**. 0.02855 (0 .299). M* (FIELD). *lH raj! (BDEN). -0.01580 (1.266). *~1Jit (METH). 0.10277 (2.599)**. - 0.44481 (2.462)**. - 0.95029 (3.81 6)**. 0.06077 (0.856). :ffi"~I1l'i(EGMY ). -0.00005 (19.098)**. - 0.00018 (13.862)**. -0.00032 (16.115)**. - 0.00003 (6.264)**. a (Scale Parameter). 1.00000. 4 .53081 (65.538)**(b). 6.25539 l.72426 (87 .704)* *(b) (l9.830)**(b). e (Shape Parameter). 1.00000. 1.00000. 0.00000. -31177.99. -17519.24. - 18002.68. Log Likelihood fi~m~§. it:. m. 5720. 5720. * * &. * ~~0gIJ~J:E. (a) t2iWPl~ t fiUH-tfil . ff ( b) t fiI!!Ji~~I*IUJiW(f]~1* 1. 0. - 200 . 3.57854 (30.563)**(b) -16415.76. 5720. 95%!Jij 9O%Z.Jli.7.I<t\Il. 5720 0.

(19) 5- {IF .Jt. -*.5 ~m "~fi. ;k -t ~JHEI' -t~!f15-tt. -4- ~r, ~J J.l!- ;k -t till f"1 -. .t. ~. Je. g, k.$. ... -t 1.. .:t..*!i. ~. : -ki1:. kjJ.l--t j:. 72/73 -f. : 9C~MrB~ (DUR). ~$5}~. M"~1t -m-~IJi (CONSTANT). exponential. 1.60641 (7.026)**(a). generalized. Wei bull. log-normal. 5.56069 (5.030)**. 8.87602 (5.776)**. 1.16899 (3 .107)**. gamma. 4~. (AGE). - 0.03494 (4.040)**. -0.27153 (6.567)**. - 0.50289 (8 .837)**. 0.02724 (1.854 )*. ~l@. (MA). 0.22349 (4.101)**. 1.18784 (4.353)**. 1.79754 (4.800)**. - 0.18905 (2.393)* *. ~{9:. (AF). ~~~H~1t(EDU). 0.19447 (5.705)"*. 0.96790 (5.760)* *. 1.68718 (7.400 )**. -0.02725 (0.521). M*(FIELD). 0.01047 (0.359). - 0.04218 (0.294). ~1H~m(B DE N). 0.06126 (0.794). 0.49798 (1.312). I. 20272 (2 .292)**. *~1Jffi (METH). 0.28867 (6.625)**. 0.36996 (1.727)*. 0.27991 (0.954). 0.18533 (2.951)**. 1ff~Iit (EGMY). -0.00005 (11.856)**. -0.00015 (7.041)**. - 0.00029 (8.858)**. - 0.00002 (3.182)**. 6.81499 (87.796)**(b). 1.43265 (l2.040)**(b). -0.22891 (U65). a: (Scale Parameter). 1.00000. 8 (Sbape Parameter). 1.00000. 1.00000. 0.00000. - 31853.17. -16915 .79. -17434.18. Log Likelihood f~. B~~JH3. a:. 4 .91172 (66.002)**(b). 5455. 5455. * * '& *. (a) jMl.\pq~ t fiUHHti . 1!f ( b) t fl!lJi~~i* ~MW IY-J~1t 1. *. - 0.04271 (0.979) - 0.13503 (1.215). 4 .69357 (28.959)**(b) - 15443.15. 5455. ~~5JIJilti,@ 95%~ 90%:L!IJi'i'7i<$. 0. - 201 . 5455 0.

(20) *6 1&:WH~ ~f{. ~.*M~ ~~ ~ ~. : ~~AAFFl'. ~$ 5t~ l'!;;:f; ~ ~ 1itJltI~. (CONSTANT). 1 ~ *~. •• ±74~. ( DUR ) generalized. ex.po nen tial. 1.42050 (6.483)* *(a). Weibull. log·no rmal. 5.90622 (5 .794)**. 10.56165 (7.103)**. 1.76902 (3.907)**. gamma. ~fi(AGE). - 0. 00582 (0.769). - 0.21 903 (6.150)**. - 0 .45644 (8. 522)**. 0.00362 (0.236). ~t@(MA). 0.53943 (12.768)**. 1.22835 (6.513)**. 1.87 119 (7 .015)**. 0.58112 (6.468)**. ~~(A F ). - 0.1 2423 (2 .112)**. 0.57979 (2.240)**. 1.66990 (4.5 96)**. 0. 081 66 (2.219)**. 0.42397 (2.553 )**. 0.73793 (3.081)**. - 0.16634 (4.731)**. -0.36983 (2.377)* *. - 0.31307 (1.426 ). - 0.19769 (2.53 7)**. 0. 1 634 (2 .853)**. 0.28475 (0.996). 0.59514 (1 .474). 0.29275 (2.023)**. - 0.04026 (0.807). -0.6421 5 (2.826)**. -1.24596 (3.923)**. - 0.213 4 (2.108)* *. - 0.00004 (2 3.250)**. - 0.0001 2 13.321)* *. -0.00029 (15.437)**. - 0.00003 (1 1 28)**. 4.41396 (56 .963)**(b). 6.37118 1.94412 (78.287)**(b) (21. 754)**(b). ~ ~ fliUJt (EDU). f4* (F IELD) ~ M a1!(BD EN ). :m-~I ii(EGMY). a (Scale Parameter). 1.00000. () (Sha pe Parameter). 1.00000. 1.00000. 0.00000. Log Likelihood @I. - 24527.00. - 14136.04. - 14595.28. 4661. 4661. a:. (a) t.5WP'l~ t fiU~f.t . ,fj. >I<. ( b) t na!!j ~~~lnUll ~IY-J~~ 1. * '&.. * :H ~51gIJ ~ j£. 0. - 202 . - 0.25802 ( 1.932)* 0.15648 (1.968)**. 3.06698 (26.524)**(b) - 13370.81. 4661 95%9iJ90%Zmi~ *ifl. 4661 0.

(21) .1:.1t ~.tf- ~1. -'.\;! $!.:k ~.Jltl r~ - t- ~ It. ~ :k ___ '" -t. 1:. ~J.1~. ~~*M~~~~#:*~*~~.±74.. ~7 ~fW.~lt. : ~~Mra~ ( DUR ). ta$ ~ ru!. JW~~~ ~~:LJi (CONSTANT). exponential 2 A 6252 (9.283)**(a). Weibull. log-nonnal. 7.48 109 (5 .677)**. 11 .99544 (6.51 4)**. genera li7.ed gamma 3.10357 (4.509)"' ~:. £f:ti (AGE). -0.03328 (3.308)**. - 0.29829 (5 .891)**. - 0 .58649 (8.297)**. --0.02945 (1.132). mWHMA). 0.29247 (4.220)**. 0.97185 (2 .849)**. 1.33 128 (2.806)**. 0 .19227 tl.0 7O). ~~lMIt(EDU ). 0.10897 (2. 748)". 0.64676 (3 .218)* *. J .18330 (4.280)**. 0.06940 (0.679). M~(FIELD). 0 .03367 (1.027). 0.01 735 (0.105). 0.05865 (0.258). 0.00872 (0.103). ~/J!ttit (BDEN). 0.22779 (2.453)*'". OA1224 (0.894). 0.66950 (1.057). 0.27837 (1.121). ~Jmttj i! (METH). - 0.11 75 5 (2.240)**. - 0.43249 (1.664)*. - 0.67174 (1 .838)*. - 03 1806 (2.503)**. ~~IR(EGMY). - 0 .00008 (20.67 6)**. - 0 .00018 (8.743)""". - 0.00029 (8.876)*'". - 0.00008 (9.932)**. 4.9803 7 (5 7.610)**(b). 2.28668 7.05373 (76.861)**(b) (22 .756)**(b). ~~(AF). a (S cale Parameter). 1_00000. e (Shape Paramete r). 1.00000. 1.00000. 0.00000. -26039.46. - 13676.23. - 14064.05. Log Likelihood ii~1iilt§. !:E:. fi. 4345. ( a) t5~Pi;5\ t tiUi!l;Hil • ~ * t fjUIi ~ m t%ti!!jtfl'f.]~~ 1. ( b). 4345. * &. '" 5 m?tgl/iNi.@ 0. - 203 . 2.97538 (23.830)"'*(b) - 13084.02. 4345 95%W 9O%ZIi'i;!* .iV1. 4345 0.

(22) '. ~Hl ~ ff W~::t.J~j§fili .fl&;, ~.lt~~:1J e"Fli:~tx I i'F ~ 1t~ ~~. ( AF ). 1i~~iiiIT~~. Z~.. , ~ Jit ft!!1M lt ~~~tlt ~IJI1'F. ".g: A z~1i- ~. 1~n4cp5t 5i IJ f=i~ ~*~~. ( AF= l ). ~fMWt!~f~.1l. 0. =1"~:Jl1M~i±:fIJ(.J~. 86.3%,& 92.7% '. ,. 0. ~ 3. EE~. '. fEJllt {fu 'H(.J ~ ~ftM rf3~~m. Jc.\ ~ :&f-r~ ~ I11~. if! AF Z. ( it7). {* ~ t!IJ ~ ftJ3. ~tf. '. 1 cp m%J!1H~ Ii~ I'(.J .lt~JtEjjIHJl. E13M!JH~~* ~IUB Ni~~. ,. ffij3(f1Jtlj ~1f!tNi. 0. :{f}Jt$I f'F::lfztt5ifJ( SEX ). 1J® '. ~1M~:m~tJl.ilIDfBiWflff; ~tt. 1'(.J~.MM.~ ~ a~ o ~ ~~~ z - ~ n:{f~ ~~~* I. ~~I~~M~~ ( tt8 )o mm~3z: j'i W~ ~1P!*~~ IJ. '. *ff;*~"'~~. •• ~~ .. ( SEX = 1 ) '~ Jlt tt ~. •• ~~~~MM ~~ M~~'H~fflM I. ~JIt:ftJlDltrfloo it*~~~~~~W~t~Ufl~~¥}?t5iIJ1mQt~t-t. ~ ft.fi. ( ED. '. , _ rJiti. , ftSIJltRlH9: ( AF) ~fi ~ IJ ( SEX ) Z rJJff;f§1l;~ z jU~~ t! ' 2lXA1* lX l'(.JliJ. rn'lH~sr~ {1!. ~jJ IJ. 0. ) i , ~ • • tE~m . McpW~~. '&J3.~ .m &o~. (@.Ei"1i.fi± ) ( EDU= l). W!IJfl.~~tEtl<:I {'F 1Jrn:i~ 15i~P.Ji#. {ili 1P~ I'(.J I.~~1f~ '{fljjJ~IJ I~ft';J~rf3'~l1 15iff~ J1 IJ. &~:tHfHl~lf.]M*. 0. ( FIELD ) , FIELD. fPHimll JiNi " Wi I". 0. ~~Wlrf3't(.J~ ., ±.~. :t£iiPI*Jl.iIm.if!'lml'ift£l.O~ Z~~. .~lH~.~ z ~ ~. ( FlELD =l) A* ~Mrf3' ~i.R. 0. ft. 0. mi. ff;m~ ~1t ~1Joo A::t.Jm~~~ ~~ ~M *z.~ ~~~o t5t ~ :£f ~w.z. ~Mrt ~ ::t.J lit*. QLOO :iJo71KtjU;;'Jlf'F ~. 0. tt~. 2. ~:£f~~~§ff. 0. HIJ ~~I1.V$:~~' fZSIJIt~~@"~ {JI;Af*'iiI~. *;z~m*~1IH,~ !:E ti:50~t M:miH~. ( BDEN ) f$~ K:~ ~. cP , :ft1'FH~IIj&;' ~Jif!mst£t ~ $~ ( BDEN= l). '. m (lfti~~Jli:j!f. ,. ,;tt~ ~ WlrB' ~ ~. , ;!tliJffE ,*lZSI z -~ t:*!tf71J*/J\ , ffijtE ~. 3. ,. r:p ,. BDEN Z~. W ~ ~ ~RJli.'~M.fi~. m&o~~ ,ft~ ~.1f~ ~~~ .m~ - $~~fi o tta~m.u. •• m 1J ~ ( ~T H). , . ~ . m~m~ ~.. W g.'@W~m&~ ~.' .~B m•• Z~.~. ~o* ~ ~ *~ ' ~8 ~~ ~~~'~. ••~Z.. , ~~tE A ~ cp ~M,~.~~~~ ft~ ~ o I ~ .~ ~*. - 204 .

(23) :L it a.tr-~~.!l!- ik.. -t.-W! f.ll-.f· i'f J!.I!!. ~~ l1:. ' 7, ¥3RE ~IE.*eliI#!rF f). J:. f(.J M1ra9. 0. *.. . .J... :i. ~£.t~. jliJ~!f!-~1:.E1 c ~l±lml3~~1HII; ~-15. ~mI~ X ':.Wa&~~M o m~ ~ ~m.a~ R .M ~~m'~S~. ••M. iRiff . a:~*,,£ . *~ ~ tl] ·. &.L .. ~ ~ ~ M'iSt~. J'lIJ ~~Rra~1I~tu{'F 8g",n%t~ ~:fJ(;.¥IJIfF. • ~IJ~~a: rB9~;j;f4fJtI{'Fit::t:~~ ·. iltlF!i ~~iJfilf.f~.tu1'Fm:;Wti. .g ao*ft~M %~ ~·~~n&~~~&~ma ~$ 5t ~IJ~. m::t:·j jr" '. •• ·. 7.8%& 9. 6% • ~ ~$ 0~ IJ ~ 2.71% & 2.44 %. 0. ft.~ ~~ ~ d. ffij. .'!tIl~~JT ~~~~~~ 4.4% • ~~~ m ~ :iIJ 2.90%. $~ j~~ra9:#.tmtt I {'F- .:g~,. 74 iFfw~mfft. ~Jfr& A .m~~~~Rl~r~t1Hl'~. ( METH=l ) .. ;!t {lf- ~~iE '. ffii 74. 0. 0. Wf,t B';:1 , 72 / 73. $ ~ ~ iJH/j. 74. ~l1t 72/ 73. 0. $~*g83 ~ Jf-J. jttt¥!JI{'F~ rB91l3ti~~. $.aIJ~0If~Ail~~~&rm lt ~~miiB1mf~I1'F. '. ,. ~ Jlt. *{~lX~. ~ o. ~~. . Sl; {r~ ?ttJT *:W.~~!l~ (J:ifn~ fflWl I ji ( EGMY ) fl;lc~Wl rJJ~~. 0. -Bffij~,ft ~ ma ~~@ ~.~~m.I~~~~ ~ ·~a. ~ .~.~~,. ~~ffij i ~~~~ ~J:..~ ~~ M~tt~ n.~~~m*~I~ ' E1~~~.MM .~. 0. &z ' ~~ I :at~lJ£1f ' ~~':fJ(;¥!J ~m=~I1'F-. 0. @tE~ 2 W~ 3. {pnlm*••• ~~~Iii ( EGMY ) .~~~&ffij~~Wl rB9~~ , .f!P aJ3JtJHW f(.J 1-f~. 0. J3.mtJ:litMOgf*ttm'i"1i:i1I • f). GGD. ~1f 7t7JIJ~ - 0 .00003. El3 ~. "& -0.00004. 0. ~ll! - tl75HJT t!5j IJ~.Jt.. )'IJ1n'iR 4 '" 5 . . 6'" t1l!1J:If'rEJ '. 7. 1iID:.~ 8,ij. 0. 0. '. •. JJ~~3Ir~. 0. •• ~~,~-.m:JtEjm & ~M.. '. 0. :fJt1M;-}7JIJ*'t~t!W"3(~.~~1Il1m2.t {i5~. *~J:.*~ 1£~. ~. - 1ilPJQHJ~~lltW.g:A1JtWJ-m&~*~ ~. ~l1tm ~ft j] ~~A~~~ ill .B1a~I ~ o. ti7 Z {Uf 9EW fWf*. EGMY. 7t!CIr.J{~~~li!J!~fi. : :JtEj~IiU~~ :1H1tt-th~1i~j]~~=1':f ' ~Il~lfl.iff • f11~~ff '. f-j Wi.g:A i~~i -. Ri ¢l , ft. , !J3tlJl~ :tct!7}5jIJfi5 tr Z&U~::t:*. 4 jij,iR 5 8'-J 72/73. iffi5rr!i~. ,. I. , ;1t~~. @ flJ ~ -~. ~~if~. ( AGE ). ~ ~~ '*.MM .~.' ~ *.~~(MA = l) Z~. :JtEj M~.o:tc~~ fi~. - 205 .

(24) ~tt~m&'~~~~z~tt~~~AAM~~'ffi*~~ttZ~~MM&~~~ o ft~~n • • ~tt~~• • • .~. ~. **~. &E•• ·~M-~OO~ •• Z.~~.'. ~~.~~~X§.I.·. ,. • • ~a~ I~·. WAm~Z*~R:t1:~(*~. ~~OOfX:Stz~tt*~*~~Iif n~~fi~~Zffi&M.~.6. '. ~.AAM.~ o ~~~~. , Dfttl~~1ItIif ' ~Jlt~~tt~III{t. 1iN~: 4'~tt~II I{'F. ••. - ~OOili. I. ~.MrEl'.-fk. 0. ~jfij. 0. , 1::. 7~~~~.,~JltfiM.~~.z.~~. ~~li**~~~~'~~~AZ~~o ~~.~ -.~£.4~.7~~~. .... ~a.~tt. •• e~~tt.~tt. [email protected]~.*~~'.ME~.~ttz~~~mm~ft~.Z~~itm ~\,.a~m,~~tt~~~~~m~~2.~Wm~*~~~,ft~.a~m~~. ."f;( {g:5., 1iN~ f±l ~ 111'$ !I~rul 'H~. 0. . :ftfr!1~~~*5cm:±!?1~~e : !!JfiW~~1VmmfJilf1m ~ . ~fr1Vm aZ.~RW*M 1. 0. ~~!t~J'!. ·scale parameter ). :g~~~~:;r~~H'.:UIJIf'Fwt ? Jlt-m~~&~{E a ( .m.~. 2.*'1 J:.ffii EI3;& 2 &. 3. '*-. .;Jt~ Weibull. 5H~c~ji 4.72142 &4.69355 • ffij log-normal 7}~ii 6.55549"& 6.71000. {II ~ 'tit ~. ti 8 (. ~~. •. It · shape parameter ) Z iJ!,i ~ , 0. Generalized. Gamma ?t~~ITf.5~*~ ~~Wm~M.,a ~lI*~ ). te*~. •. ar~¥1j 1.59329. "& 2.11070'. · ffi)J1tiln~IJIf'F.. ffi1JD. : ~~Mra'Jiltft(fJ~~~JIff"J!~zH!ff.Ja~JIf'F. *EI. {lI.{J)Ii~I'f.J~M 1 0. •••• a*Ml~·.~~._M.~mH~*~B ( ~. f.J.ta$ ( .F!n m:~~) ~*-. Wlrp,±\j1m '. ~3f~if!±. 0. 0. ~gZ. '. ~1Mf.Jfi~. - 1JOOjj*-$!IH~~~~. J:ff.J~1.J.~*' jfijJ3.iiJtmfl;t~J:(fJiUtt!l*~Q1JD. · Q.tfi1mJttt~.~ ( tt9). lFH~lEttmtAI li?t!Cz,1iW-J. • 1;:2 jfijC!f{flifu 1M IY-J i*"\j Iit JHXlrjIf'F zM. ~.rl¥Hi-WJ:t~~8JtID!rzt.5.. Weibull,HGaq!ftf~~~, ffij Weibull. 0. 0. ~-1JiID. ·. . ~fitH~Hfu. 1ZSlJij: {th 1M&'~~fJi;!t(~iHrlIit. ;& 8. 0. IiJM exponen tial. 0. 7J'~~. 1U log-normal X;;:e: generalized gamma. - 206 .

(25) :r...{t ;fi1f~1'! ~ ~ 1'M r~-t· ~ Jt.~;k-4-*-t ~u:_ ~~. ~8. ~ ~ ~ ~T ~~MM ~ # ~ ~~~~:~ft.t~. exponential. generalized log-nonnal. Weibull. gamma. 72/73 iF{5IT~~ 0:. (scal e parameter). 4.72124. 8 (shape param eter). 1. Log Likelihood 1@:. - 63060. 15. Lik elihood Ra tio. ~. H~1ttlX §. 1it!J!J l!l!fli~ 1@:It §. 623 13.26*. - 34463 .98. 6.555 49. 1.593 29. 0. 4.0493 1. - 35488 .05. - 3 1903. 52. 5120.92*. 71 69.06 *. ) 1175. II 175. 11 175. 11 175. 489. 489. 489. 489. 4 69355. 6.71000. 2 . 11070. 1. 1. 0. 3.023 ) 1. - 5062 8.95. - 27837 . 12. - 28683 .79. - 26498.63. 74 iFf58t*6 ~ 0:. (scale pa rameter). 8 (shap e param eter) Log Likelihood. fit. Likelih ood Ratio {[! B.~~It §. lIU!] IWfIB~ {lit § ti:. ( *). 482 60.64*. 2676 .98*. 43 70.32*. 9006. 9006. 9006. 9006. 529. 52 9. 529. 529. Jl:l: iJit Likelihood rati oIiU IIJ;) GGD. i.:\iIS~;t. ITii m. 0. JttjJ exponential lHlfl'iiIii fWllatlJ~. I!fZ GGD tt-.J~rJI; 7t~ • ~ 1t likel ihood ratio test TI~ X' ( 2) Z 7tliC • ITii X'... ( 2 ) Z fi«~ 5.99. 0. Weib ulJ !Ii! log- normal JjIJ7tlJ ll1.HHf -@l lla ltillfli {!f 'f GGD tr..J~rJI;7t~ • Jt. likelihood ratio tes t 1fJ ~ X' (])z 1i~ • ffii X'.... (l )Z fl~ 3.84 likel ihood ratio 1JJi~1U~ O • * HII~7ti£~ GG Dfl!!1i~~Dll. 7\. • rtkOJ~)Eexponen ti al W Weib ull. &.exponential W il!U~~lJnt fPll~tE JIt~7IJ t:Jj. 0. - 207. 0. 0. J1t £fl'. * 11f~~tl. X • ~ 7 ~ JE~ GGD z~J4. log-n or m allZ~~. 0. ~i!t~II1~.

(26) e1!f.f9$~~. , EI?I.\ ex ponential. ~ 'J\ M' Weibull ,. ffii Weibull. to. 7HicTB1 f} * ~ fJJ{I5§t f@:. log-normal. Clog likelihood Jill ). tt-J :I&* :fret1tJ, Jill x:. fr '1\M' generalized. gamma ° 1Zfr~ 75£1. J'& m1Ult$t~~ C likelihood ratio test) C §l10) ~ ~m Weibull :5T~~Mtf l¥J lt exponentia l tit '. ffiillt!1lt log-normal :5T@Ctif ' ffiill. gener alized gamma :5TWC x:.~ 1t Weibull :5T m:* f~tif ° lIt-*6~:f§-i" ~t;:E. H 72/73 £FQt 74. if~ .*¥. '. ~0 ~IJ0,:'!Jjt!tlit5ct! !IHf~~{I5§t* ~. , ~~iiU. , 1J(:fr~:ffl3~. ~m ~~ ~ o~-~OO,1J(: M. mm • • ~ a~~~51~~fim a *tt-J~ ft ,~~ ~. 8.:L72/73 iftjJ ,. ~fI51~Ttt-J a51~IJ~. 1,4.72142, 6.55549,. to 1.59329. °. ~~~ ati ~ ~~~ :5T~ T~m.* ~•• ~~,~lIt~~~~.~~:5T~*~~. *•••. ~~~. MOO ~ '~~. l'&~ ';ff~ 9 ~ ~. 0 •• C§lll)°. 10 cp , i!tfr~0~IJJIJIfj~ tt~~ ~~~t! • •. e. ~ ~ a~. . o ~ *6 .* ~~w . 8 m a -~,mfi-~.~~A ~ . ~~~~9CP'~t!. B1 a f@t£ 72/73 ifEz 74. £F51~IJ ~ 1. 72426. ~t':E B1 a 1@:~p;p:j*Jl~ i$ cpti]J f'f;f§a * tt-J~~. fD. 1. 94412. '. m'i'~ili. ° ffii~ 10 cp. ,. , [email protected] 1.43265 fD 2.28668 0:9:t!. • • ~a • • ~m;:E ~ RW ~ ~ .~ ~ ~ *.m ~ fi~ ,~Ifj ~. ~.,tt~*m. ~~o~ ~ ~.~ ~A~ ~- *~ ~I¥J ~~.:L -o. 4.. *••• *~.~ -. ~ ~. ~. ~*.mAh~~~*~'K~.~~~ ~ m~ft fr. •••. •• m.o. ~I~ • •• ~'m~I~~8d*&OOti~I.51~~M~. tt~I ~ ~._&~.AA M o~m r~Mm • • ~ • • ~U.~~.~ J. if& 74 iFJf*'l-&ffm5t;ffi-i-i ( s uvival analysis) ,. :ft1PHlmM:~IJ. 72 / n. .. M* ... *~. h~'&fflAAI~~ ~~ . AA OO~~ ~:ffl3~~ g ~ ~. o~lt~~fI~$51~. expo nentia l , Weib ull , log- normal & generalized gamma ~51 ~. - 208 -. ,. ~ ir~. «.

(27) .:c.1'f:IX. -4 .~ ~ I!-!k 1" ill. ~9. r.,- -t.. ~ J,I:!.d~ *.t- ... -f. ±. ·:t.J;Ht. ~ ~ ~ ~T~. MM##~~~ ~.:~~ •• ~. expo nential. generalized Weibull. log-normal. gamma. 72 / 73 fF.{3gt~ ~ 4.53081. ex (scale param eter). 6.25539. 1.72426. 0. 3.57854. -18002 .68. - 16415 .76. e (shape paramerater) Log Likelihood ~ Likelihood Ra tio. ~. ~~ ~ti § t/iJJ lii fN~ ~ lfl: §. - 311 77 .99 29524.46*. - 17519.24 2206.96*. 3J 73.84*. 5720. 5720. 5720. 5720. 242. 242. 242. 242. 74 fF.{3;IT~m ex (scale parameter). 4.41396. 6.37 118. 1.944 12. e (shape parameter). 1. 0. 3.06698. - 14 136.04. -14595 .28. - l 3370.81. Log Likelihood ~ Likelihood Rat io fig ti~~lfl:§. 1&: l}!Jlifi~1W:~§. IT: (. * ) Fcl :&. 8. - 24527.00 22 31 2 .38*. 1530.46*. 2448 .94 *. 4661. 4661. 4661. 4661. 278. 278. 278. 278. 0. - 209 .

(28) exponential. WeibulJ. log-normal. generabzed gamma. 72/73 iF'f5 rr~~ 4 .9 1172. a (scale paramet r). o (shape parameter) it. - 31853.17 32820.04* Likelihoo Ratio it.. Log Likelihood B~it.li§. ~~Iif.~fi~ §. - 169 15.79. - 17434. ) 8. - 15443. 15. 5455. 545 5. 545 5. 247. 247. 247. 247. 4.9803 7. 7.053 73. 2.28668. 0. 2. 97538. - 14064.05. - 13084.02. fIi - 26039.46 259 10.88* L ikelihood Ratio 1l.. I'iil ~ 8. 4. 6935 7. 5455. Lo Likelihood. 11 : ( "' ). 0 3982.02*. e (shape parameter). 'flt!Ylerft~1@:fI §. 1.43265. 2945.:::8*. a (sca le parameter). .~iiIH3. 6.81 499. - 13676.23 118 4.42 *. 1960. 06*. 4345. 4345. 4345. 4345. 251. 25 1. 25 1. 251. 0. - 210 .

(29) ~~~~~~~~~MM -- ~~~g*~~*~~~~. -mfr~lHHllfiR~. • mt~~lt ( scale parametera). a.grM~lYl:k~ 1. I. ~~*~. •• ~tt~I~~ •• ~.~.M~.~ ~ m~'~A~~~~~~~O~ ( time dependence). 0. ~9C.W=JrFfl4i!ffl1m-1ijJ. lIt*5JfaJEl:r#il~. ( 1992. ) ag~~JB~m~. , .~1*~Ijt~r~ O.32%iIJ 0 .7%. B~rf~~OJlt~m~tt!IJIt'Ft't-J~l¥. 0. ~-l1iID. · fr~f5J~WCZl'. 0. · 11JJ*~"'* ~.tltf*~Iit. aflirr{i!~:,JHElIi". *'~~~Aff~.~~*~~It'F~~ff~~~~m~~~o If'F~t§il~Wtf1Jtf*"\WIit (reservation wage ) tr.Jfi!lgtm~~~Jl.,. ~*~~~~~A~M·@~~-~o~-~OO ' *x~m~~ f¥ tj:!('l'Jmt~~l*i. affi'i"fiJE '. Jl.,il3!±1~ilJmtJl~1!M. '. @~M:t't-J. 0. *X. •• ~It'F~~~. lItm~1iPJ~n~M:~~~~. alHtHtk '. fi~~~~-~lUf~. ,. Afl ' rna?' 72/73 1:FW 74. -fFpw~Jm. rl'J' , lI~j@ ~~~~J!1Jttr~JlIj?'!~ , ~~3t{3ITf*fiffm~jJj[@1t55-tcp!J)\~ilJm~. * tffii* ffl CJ. ; fi. ~~JE ft-J Nii-fF it ~. *. J:t~. , ft1f ;!ifh!t ]I! ,jU~ ijR:h. tt. lift 1 .. 0. EI3 rliU¥l!Y-Jlt~mI~f.!i!:lrii]~ : III Utl<:If'J01iX* c ffl1JIl· tJF:{iIiiR.·Hl; (2) :5':F.f~- Ij1f7t1e h ( w) • ti h (w) :J:IHI'J-Wi4l~t: f ( w) ( first order stochastic d ominance) . JlIJ R ( c;h) INn R ( c;f ) 1DJ~~ - m.«'z~.nlt~§mH6~PJ#J! 0. Li ppmand and McCall (1981) m215]!{f!J2221izt-t~ 2. . *,:c jjfz :B:JIt~$t;Jt~J:t01J ffl:~.If.i. 3. . !lt1#!~M1t~J'!'. ( proportional hazard rate ). ftiFlli&-ffJ:U2!iiSJlt. 0. 0. J1 = O . I:iU~11'1t a=l o. 4 . ~:tE.~AArJ.lrm~£.(iVl· ~ft9.;~lUlra~B'@ll,IJjIf;. fin. 0. ~H;l!/j ffii e:Jt-t~PJ~~ Miller ( 1981). . QIJ.ilim1i:I5~Ui ( left. J$=.. 5. ftx:Jf:l:~* I±llY:JlIN{);J.lli!i~~mHt:J1\lfl:1i?jj1X (partial likelihood funtion). and Prentice ( 1980 ) . f[J~. ( 1988). 0. ~lf:!m Kalbfleisch. 0. 6 . :f;jlJSJJltlD'J§Z~*'Bpq~PJ~M!~U~:*I*:I*13;.. 7. 1i:a~lijm:~*miAm:I±ltr\J~fl. censored ) tr.J1\!~. 0. ( 1988) ". 0. 8 . ~.L , ~~~~~~-~~~.~aM:tE*~·&~I~o~~A*m~~~9.;*AAM~~ o ~. - 211 .

(30) iI .' ~Aitft&.n1H':Et.t~I{'F. ' ~. 1iN~tl:l~ihm~tf'. ftSlJtt;r;:eh?i1f*::X:Jm.$. 0. g. ;j;fjI3.J& (1992 ) ~if!J!~~:*:l1P'~~~~Mrdlfij±ft~-mlfl ' :jt.Jl~i1Ij{ (real reservat ion. wage ) *r~ 10. . o.32%ilJ. 0.7% ' ftSlJlt$'±fit1JD:Jttt¥11If'FZ1tll¥. 0. ':J'i::1f1!HJ;Lfti= j!Z-*1t.rrMI~), =- 21og ~ ) jtl:jJ 1.. :&~~:!iIfftt-J1INfJtJ,fitl, L. ~jFilfl~ffz1lN©.fitl · ffii ,1,-X' Cn). , n ~~*ll ~{4=zJX§. JIJ ), = 68800 .76 .:kJit X'....(1) = 6 .63 ' 'Jl'1:.'. 11 .. EB~afO eJ!flf~iqj71!!2ffiif;fl§,&,*~. 1t o l7IJ1m'. **1§.. ~~Wi11flJlii~~. ~. 3 $ ED U Z{,:fIXt££9. '269-310. ./ **ill& ( 1992 ) , •. 0. W·. ~Jlt1r5fitiJJaq~1S!~ZJ%.ifHH§1it*aq~. 711f'.r 71 }jIJ ~. 0.11606,0 .72468, 1.18670 , :& - 0 .00982 0. ( 1988 ) ' "11,'U.U~~~~rlr.:Hl:JM11it~»UHE' ~~. 0. • ft1Jl~~. 121. 72/73 ~tt-J e ponentiaJ CL= - 63060.15 ) f[] Wei bull (Lu=-34463 .98) ~WlJ. N. cpI!;U~~. *~.ta3C.' .g~ t. :. $~~. 0. ilti1llrg*_ .~IjtHUIil~~$lIJln.'llUJf*zliJf~'" • • ilHl~tIJ ' RIIMtl±l. 11&0 ~!iEs. ~~tEQ~.ff~J:.zml#t , ,, f!l!j;!f~3C.'flJ ' 16 (3)' 415 -438 0. (1980 ) • "~!tt1JDI I±1D rg~I~~~zTilf1E '" @Jfia?2: 'fll' 9 (2)' 25-56 0. ~.~E. C1981 ) ,. I 1*1xh$ ( 1988). 229-298 ~!iEE. . ":WJ:. " *IU.$IY-J~~~* : !Yl~. ( 1983) ,. @iJ!H~3t.fll. "?!itlt1JD.Il±lOIl&*.IJlfUa$ZMlUf~ : ~~. • 11 ' 33-60. J ~~, i* . #'ilU~ ( 1988) • • ..l. iitJ1:H! • #eiE ( 1991) , TiJf~pftl:jJw.A.xn ~m~. ?l!tr-J :1HHWl$)E'". ijif:.~.tiB '29. '. 0. . ~1lJI1J! ( 1991). ;k:q)f~~~wA ::x:iitt. 61. ~~~I!!l. 67 iFl¥J.m71tT ' ". 0. .~~.II;I:iJl.Illli:_Hs·. ii':it :. !f:1.~Q~1!lt 0. " .1l!ti1l!m;~I.iTrtttJz tilf~· " ~ilrmfl~~nt'"H~~t-t*. '. t':ft : ¢9<:. flI~Till~PJr o. , ". iIJ~lIJ1J$Ii:*Wlr~It.JTill~· " ~lfnf]tIJt.~jlfg~~i'ta. .. i3'~t. : t:jJ. f4~TiJf~PJTo. Barn es , W. . . (1 975), "lob Sear h Models, the Duration. f Unemployment , and the Asking. Wage," Journal of Human Resources, Spring. 230·240. Crosslin, R . L. and D. W. Stevens, 1977) , ·'Th . Asking Wage-Duration of Unemploy me nt Reltltion RevisI ted," So urthern Econumic Ja umal. 43 (3) 1290 - J 302 .. 'hr nberg, R. G. and R. L Oaxaca (I976), "Unemploym nt Insurance, Duration of Unem plo yme nt , and Subsequen t Wage Gain," American Economic Review , 66 754-766. Heckman, J. J. (1978), " imple St tistical Models or Discrete unel Dala Developed and ApplJed to Test the Hypothesis of True Stnt De enden e against the Hypothesis of Spurious Slate Dependen e ," Annals de L 'filS e, 30/3 1, 227 -270 , Paris .. -21 2 .

(31) .:!... {t tt Jf. {~~ "1l! $I-. t<.. -iI';,ItJj rJI -.t- ~ )/:.. ~ k ..$...f 1::t. ~ .@. $-. Heckman , J J . an d G. Borjas (1980) , "Does Unemployment Cause Future Unemploym ent ? Definition. Question s. and Answer fro m a Contin uous Time Model of Heterogeneity and State Depe ndence," Ecollomica , 47 . 247-283. Kalb Dci sch. J. D. and R. L. Prent ice (19 80) , Th e Statistical Allalvsis of Fa ilure Time Data . New York : Jo hn Wiley & Sons Inc. Kasper, H. ( 1967), 'The Asking Price of Labor and the Duration of L nemploymcnt ." Re Fiew of Economics and S tatistics , 49 ( 2). J 65 -J 72. )(jefer. N. M. (1988), "Eco nom ic Duration Data and Haza rd Function. " Journa.l oj Fcollomic. •. Literature, 26 (3 ),646 -679 . Kie fer , N. M . an d G. R. Newmann ( 1979), "An Empiri cal Job Sea rch Model with a 1 est of. Constant Reservati on-Wage Hypo thesis," Juurnal of Political Ecollo my , 87( I ). 89-108.. Kiefe r, N . M. an d G . R_ Newman n (J 981), "indi vidual Effec ts in a Nonlin ear Model: Ex plicit. Treatme nt of Heterogc nlty in the Empirica l J ob-Sea rch MadeJ: ' L'cOl/ol1lctrica. 49 (4 ). % 5-979. Lancas ter , T. (1 979), "Econo metric Methods fo r the Dura tio n of Unempl oyme nt :' Econo m e trica , 47 0 ), 939 -956. anc3st er, T. ( 1985 ), "G eneralized Res idua ls and Heterogenous Duration Mo del: With. Applica ti on to the Weibull Distrib u lion, " journal of Eco nometrics , Annals, 28 , 155 -169.. Lan cJste r. T. (1 990), The Econometric A nalvsis of Trall sit ion Data. N.Y.: Ca mbridge Univer. si ty Press . Lippman. S. A. and J. J . McCall (1976a), "The Econom ic s of J ob Search: A Survey , ParI 1," Economic fll quily , J4 (2) ,155 - 190.. Lippman. S. A. and J . J. McCall ( 1976b), "TIle Economics of J ob Sea rch: A Survey , Part II ," Eco nomic Illquiry , 14 (2) , 300 -317 . l ippma n, S. A. and J. J. McCall (1981) , "The Eco norn.ics of Uncertainty ," in K. J . Arrow and MD . In triJjgator cd. Handbook ofMa th ematic Ecol/omics, I ( G) . 21 1-2 84. McCal l, J . J. (J 965), ''The Economics of In formation an d Optimal Stopping Rul es," j ournal oIBusiness . :100- 317. McCall , J . J. ( J 970), "Economics of In formation and .J ob Search," Quarterly j ournal of Eco no mics, 84 (1). 133 -1 26. Miller , R. C . ( 198 1), Survival A na(vsis, Ne w York: Joh n Wil ey & Sons In c.. Mortesen. D. T . ( 1986). "Jo b Search and La bor Market Analy sis," in O. Ashenfe lter and R. Layard , cd. Handbook of Labor Econom ics. :2, 847-9 L . Sh arma, S. (1 985), "La bor Market Hi stories:' Ph . D. Disse rtat ion, Department of Eco nomics, Cornell University , Stigler, G . J. (1961 ), 'The Economics of Information." Journal of Po litical EcolZomy , 69 (2) , 213-22 5. Stigler. G. J. (1962), "in form al ion in the Labor Market," Journal of Political Econo my , 70(1 ), 94-1 04 .. - 2 13 .

(32) Toikka , R. S. (1972), " Supply Responses of the Unemployed : A Probability Model of Reem ployment." Unpublished Ph .D. Dissertation, University of Wisconsin .. - 214 .

(33) .I..1t~t ~ ,f~~ -lI!- ~ -t J\Ij. f·1 - ·t- )If J.e. g *- ~.f'" jc:Z .~ $-. JOB SEARCH MODEL AND UNEMPLOYMENT DURATION - - - THE EXPERIENCE OF COLLEGE GRADUATES IN TAIWAN. Chu-Chia Lin Department of Economi cs, Na tional Chengchi University. ABSTRACT This paper, firstly, build a simple job search model. to explain how search cost and wage distribution intluence the hazard rate and unemployment rate . Using the "Survey of employ ment Conditions of College Graduates in Taiwan, 1983/ 1984 and 19 85," and survival analysis, this paper finds that sex , field , job-search method , and expected wage have sigIlificant effects on unemployment duration.. AppJying exponential, Weibull , log-n ormal , and generalized. gamma di_stributions, we find that estimated scale parameters are always greate r than unity . This res ult implies that the hazard rate in crea ses as the unemployment duration extends. In other words, there is a time dependence. In the meantime, the est imated scale parameters arc very sensitive on various distributions, which implies that one ha s to be careful when he chooses an appropriate distribution to de scribe job search behavior.. - 215 .

(34)