Final Exam - Solutions

Size: px
Start display at page:

Download "Final Exam - Solutions"

Transcription

1 EC 70 - Math for Economsts Samson Alva Department of Economcs, Boston College December 13, 011 Fnal Exam - Solutons 1. Job Search (a) The agent s utlty functon has no explct dependence on tme, except through the exponental dscount factor, and nether the wage offer process nor the annuty process has explct tme dependence. Thus, gven that there s no means to save and that annuty s constant every perod, the only thng that vares from one perod to the next s the most recent job offer held by the agent. If n two dfferent perods the agent held an offer for some amount w, then from the agent s perspectve, the futures startng from each of those perods are equvalent to the extent that the probablty of any future path of all varables s the same f the same contngent plan s used. Ths statonarty s the reason why an agent that has chosen to work at some job wll never fnd t optmal to qut the job and search for a new one at some later date. Consder the agent s decson when the job offer was frst receved. Gven that the choce was optmal n ths perod, then at some later date, when the agent s consderng whether to contnue to work at the job or search for another job, the agent realzes that nothng has changed about her nformaton about wage offers. Her current offer s the same as when she frst began her job, and the wage offer dstrbuton s the same as well. Moreover, because she cannot save, and because the annuty s constant, nothng about her decson problem has changed, and so, gven that t was optmal to start workng at her current job, t s optmal for her to contnue to work at her current job. (b) The above argument about never quttng a job once startng t mples that the agent has a threshold wage offer, above whch she would accept the offer and stay at the job forever, effectvely stoppng the job search process, and below whch she rejects the offer and contnues to search. Let ths threshold wage be denoted by. Denotng by ŵ the most recent wage offer, the Bellman for the decson problem s: V (ŵ t ) = max e t (y 0 + e t ŵ + β E[V (ŵ t+1 )]) where ŵ t+1 = e t ŵ t + (1 e t )w t. Keep n mnd that w t s only observed f e t = 0, and so s a random varable at tme t, when the agent s decdng whether to work or search. Also, f the agent searches and receves offer w t, the agent cannot work at ths job untl the next perod, and so receves no wage n the perod she s searchng. 1

2 Gven the prevous dscusson that the agent wll choose to work f the most recent wage offer s greater than, and choose to search f the most recent wage offer s below, we know f the most recent wage offer s equal to, the agent s ndfferent to searchng and to workng. If the agent choose to work, the agent s lfetme utlty s just the present-dscounted value of the stream of ncome y 0 +. Gven the dscount factor of β, ths lfetme utlty level s y 0+. If the agent 1 β chooses to search, the agent receves y 0 n utlty n the perod of search, and the dscounted expected value of the offer receved from today s search. Thus, the value of search f the current wage offer s s y 0 + β E[V (w t )], where w t s drawn..d from a dstrbuton wth c.d.f. H. So, when the current wage offer s, the agent must be ndfferent, mplyng: y β = y 0 + βe[v (w t )], (1) where w t, the newly-receved wage offer, s dstrbuted accordng to H. Now, usng the dea that the agent wll accept any offer w t that s greater than, whch occurs wth probablty 1 H( ), we know that V (w t ) = y 0+w t for any w 1 β t, snce the agent wll receve ncome stream y 0 + w t forever. On the other hand, f the agent receves a wage offer w t <, the agent wll reject the offer and search agan. Thus, f w t <, whch occurs wth probablty H( ), V (w t ) = y 0 + β E[V (w t+1 )], where w t+1 s the next wage offer. However, E[V (w t+1 )] = E[V (w t )], gven our assumptons about the wage offer process. Thus, assumng the p.d.f. assocated wth H exsts, E[V (w)] = 0 V (w)h(w)dw = H( )(y 0 +β E[V (w)])+ Rearrangng equaton (), we obtan (1 βh( )) E[V (w)] = H( )y 0 + Rearrangng equaton (1), we obtan (y 0 + w) 1 β h(w)dw. β(1 β) E[V (w)] = (y 0 + ) (1 β)y 0. Puttng these results together to elmnate E[V ( )], we get (1 βh( ))(y 0 + (1 β)y 0 ) = β(1 β)h( )y 0 + β (y 0 + w) h(w)dw () 1 β (y 0 + w)h(w)dw. The frst term n the ntegral evaluates to βy 0 (1 H( )), and so the above equaton becomes (1 βh( )) +β(1 βh( ))y 0 = β(1 β)h( )y 0 +β(1 H( ))y 0 +β wh(w)dw.

3 All the terms nvolvng y 0 cancel out, leavng the equaton Ths smplfes to (1 βh( )) =β =β =β (1 β) = β wh(w)dw (w )h(w)dw + β h(w)dw (w )h(w)dw + β(1 H( )). (w )h(w)dw. Notce that the left- and rght-hand sdes of the equaton are functons of. The left-hand sde s an ncreasng functon, startng at a value of 0 when = 0. The rght-hand sde s a decreasng functon of snce the lower lmt of the ntegral s, and snce the term n parentheses s always postve but decreasng n for any value of w wthn the ntegraton lmts. Moreover, the rght-hand sde has value 0 when = w. Thus, by the ntermedate value theorem, there must exst some that solves the equaton f the cumulatve dstrbuton functon H s absolutely contnuous (and so the ntegral s contnuous). Ths equaton mplctly determnes the threshold wage that the agent uses to decde whether to accept or reject a wage offer. If we assume that wage offers are unformly dstrbuted on [0, w], then h(w) = 1 w, and the above equaton smplfes to whch yelds a quadratc equaton n, The two roots of ths quadratc are (1 β) = β w (w ), β w + βw = 0. = w β (1 ± 1 β ), of whch only the negatve root makes sense (.e. the postve root yelds > w). Thus, the threshold wage = w β (1 1 β ). (c) Clearly, the threshold wage s ndependent of y 0.e. y 0 = 0. The ntuton s qute smple. Our agent receves the annuty no matter what she does. Snce she has no way to earn nterest ncome va a savngs opton, she has to consume the annuty n the perod t s receved. Moreover, she has lnear utlty and so her decson about whether to begn workng or not, gven her level of consumpton guaranteed by her annuty ncome, s at the margn the same no matter the level 3

4 of y 0. Ths would not be true f she had, say, a concave utlty functon. Then, as her annuty amount ncreases, the margnal value of workng and earnng wage ncome ths perod decreases. So, the opportunty cost of searchng decreases. Moreover, at hgher consumpton levels, (most) concave utlty functons appear more lnear for a gven spread n consumpton, and so the dsutlty from the rsk assocated wth searchng decreases. Ths reasonng would suggest that hgher levels of annuty should nduce our agent to ncrease the threshold wage for a gven utlty functon. On the other hand, a smlar lne of reasonng suggests that ncreasng the concavty of the utlty functon for a gven level of annuty should reduce the threshold wage. For a general utlty functon, the threshold wage can be shown to be the soluton to the followng equaton, usng an analyss smlar to the one conducted above: (1 β) [u(y 0 + ) u(y 0 )] = β [u(y 0 + w) u(y 0 + )] h(w)dw. In ths case, after applyng the Implct Functon Theorem and smplfyng, one can obtan = 1 + (1 β)u (y 0 ) + β w u (y 0 + w)h(w)dw y 0 (1 β)u (y 0 + ) + β u (y 0 + )h(w)dw. The fractonal term s postve, but the magntude s not apparent.. Managng a Hydroelectrc Power Plant (a) The current-value Hamltonan s H(L, z, µ) ALz + µ(r z) + λ 1 z λ (z Z), whch takes nto account the constrants on the control. Accordng to the maxmum prncple, at the optmum the paths of the control, the state and the costate satsfy: H z = AL(t) µ(t) + λ 1 (t) λ (t) = 0 λ 1 (t) 0, z(t) 0, λ 1 z(t) = 0 λ (t) 0, z(t) Z, λ (Z z(t)) = 0 H L = Az(t) = ρµ(t) µ(t) H µ = R(t) z(t) = L(t) lm t e ρt µ(t) = 0, where the last condton s a transversalty condton ensurng the shadow value of the state/stock varable L (n present-value terms) approaches 0. Ths condton makes sense, snce havng a postve shadow value mples that the rate of outflow could have been ncreased, whch would mprove the lfe-tme payoff. Ths form of the transversalty s present, rather than one nvolvng L drectly, snce we don t have to be concerned wth a non-negatvty condton on L. That L appears n the objectve mples that L wll never be negatve at an optmum. 4

5 (b) Snce R(t) = R, a constant, the requrement for steady-state s that L(t) = 0 and µ(t) = 0 for all t. These two condtons mply that z(t) = R (assume R (0, Z), otherwse the maxmum rate of outflow s less the the rate of nflow, and the lake wll grow wthout bound), and ρµ(t) = Az(t) = AR. Fnally, gven that R (0, Z), z(t) avods the boundares of ts constrant set, and so λ 1 (t) = λ (t) = 0, mplyng µ(t) = AL(t). Thus, L(t) = R. In summary, n the ρ steady-state, the control, state, and costate have values z = R, L = R ρ, µ = AR ρ. (c) Lnearty of the control n the Hamltonan mples that the optmal control s dscontnuous, takng values at the boundares when not n steady-state. Ths type of soluton s called a bang-bang soluton, because the control does not take on values n the nteror of the admssble set, except perhaps n steady-state. If the steady-state path s ncluded n the optmal soluton, then the value of z(0) depends on whether the level of water L(0) s above or below the steady-state. If t s below, then z(0) equal 0 untl the level of water reaches the steady-state, whch occurs at some tme τ, after whch tme z(t) = R. Smlarly, f L(0) s greater than the steady-state level, then z(0) = Z untl some tme τ, causng the level of water to fall untl t reaches the steady-state level, at whch pont z(t) = R. Unlke n, say, the optmal growth problem, steady-state can be reached n a fnte tme, and the control s pecewse contnuous, actually pecewse constant. Thus, the only queston remanng s whether ths soluton of gettng to the steadystate and then stayng there s n fact the optmal soluton. The only other canddate soluton s one where z keeps swtchng between the two extremes of 0 and Z. 3. Durable Good Monopolst wth Myopc Consumers (a) A consumer buys the product f and only f hs valuaton v p. Each consumer draws hs valuaton randomly from the unform dstrbuton, and so the probablty that v p s 1 p. Snce each consumer s valuaton s drawn ndependently and dentcally from the unform dstrbuton, and we have a contnuum of consumers, the dstrbuton of valuatons n the market place ends beng exactly the dstrbuton from whch the valuatons are drawn. Thus, 1 p s also the market demand.e. q 1 (p) = 1 p. (b) Frst perod proft s p 1 (1 p 1 ) and so the proft-maxmzng prce s p 1 = 1, and the proft level s 1 4. (c) All consumers wth valuatons above p 1 wll buy the good n the frst perod and leave, and so the hghest valuaton consumer left n the market has valuaton p 1. Also, 1 p 1 consumers buy n the frst perod, and so the resdual demand n perod s q (p ) = p 1 p. Takng p 1 as gven, the proft-maxmzng prce s p 1. The perod proft s p

6 (d) Thus, the present dscounted value of proft n perod 1, π = p 1 (1 p 1 ) + β p 1 4. The frst-order condton yelds 1 p 1 + β p 1 = 0, whch yelds p 1 = 4 β. The dscounted proft level s 1 4 β. (e) Settng a prce n some perod that s hgher than the prce n some prevous perod wll result n zero sales, because every consumer who mght have purchased the product ths perod wll have already exted the market. Thus, prces must (weakly) decrease n order to have any chance to makng any sales. (f) At tme t, every consumer whose valuaton s above the lowest prce to date wll have exted the market.e. the only consumers remanng n the market are those whose valuatons are less than or equal to mn t {p 1,..., p t 1 }. Snce we know prces must decrease, we can conclude that mn t {p 1,..., p t 1 } = p t 1. Thus, at tme t, every consumer wth valuaton below p t 1 remans. Thus, at tme t, demand q t = p t 1 p t. (g) The monopolst s problem s max {p t} T t=1 T β t 1 p t (p t 1 p t ), t=1 where p 0 s defned to be 1. It should be apparent that boundary solutons wll not occur, so ths s just an unconstraned maxmzaton problem wth T choce varables {p 1,..., p T }. The frst order condton for the prce at a generc tme perod t < T s and the frst order condton for p T s These smplfy to β t 1 ((p t 1 p t ) p t ) + β t p t+1 = 0 β T 1 ((p T 1 p T ) p T ) = 0. p t 1 p t + βp t+1 = 0, t {1,..., T 1} p T 1 p T = 0. Ths second-order dfference equaton determnes the optmal prcng polcy of the monopolst. (h) Start wth the last perod. We see that p T = p T 1. Next, we have that p T p T 1 + βp T = 0, whch mples p T 1 = p T +βp T. Substtutng our soluton for 6

7 p T and rearrangng yelds p T 1 = 4 p T 4 β = γ 1 p T, where γ β Next, notce that we have p t = p t 1+βp t+1 for any t < T. Let s prove that the p dfference equatons yeld p t = γ t 1 T t for any t > 0, where γ T t s some constant that depends on t. Notce that we have p T = p T 1, so γ 0 = 1. We also know that γ 1 = 4. Now, suppose our conjecture holds for every tme perod 4 β t p {t + 1,..., T }. Then, gven that p t+1 = γ t T (t+1), we can substtute for p t+1 n the dfference equaton p t = p t 1+βp t+1 to get p t = p βγ T (t+1) t 1+ p t, whch mples that p t = 4 4 βγ T (t+1) p t 1. Thus, we have that γ T t = 4 4 βγ T (t+1), whch s a dfference equaton that defnes the sequence {γ s } T s=0 1 p. Now, p 1 = γ 0 T 1 More generally, p t = 1 t = γ T 1 t γ s. s=1 and p = γt p 1 = γ T 1 γ T. Startng wth γ 0 = 1, one can show that γ s+1 > γ s for all s 0. Moreover, γs < 1 for all s 0. Thus, we see that prces do n fact decrease over tme. Also, notce that γ s+1 β = 4γ s (4 βγ s ) = γ s 4 γ s+1 > 0 for all s 0. Thus, each term n the product comprsng p t s ncreasng n β, and so gven a tme perod t, p t s ncreasng n β. Thus, a more patent monopolst charges a hgher prce that a less patent monopolst n every perod. Also, as the number of tme perods ncreases, the prce n any partcular tme perod ncreases; to see ths, notce that p 1 = γ T 1 s ncreasng n T snce γ T 1 s ncreasng n T, as dscussed earler. A smlar argument works for every subsequent prce. Thus, as the decson horzon s extended, the monopolst chooses a sequence of prces that decreases more slowly. In fact, notce that the recursve equaton for γ s must converge so that lm s (γ s+1 γ s ) = 0. To fnd the lmtng γ, solve γ = 4 4 βγ. We get a quadratc equaton n γ, βγ 4γ + 4, whch yelds the soluton γ = β (1 ± 1 β), of whch only the negatve root makes sense, snce we need γ < 1 for the prce equaton to yeld a prce between 0 and 1. Thus, as the tme horzon T extends, the rato p t+1 p t s ncreasngly well approxmated by γ, wth the approxmaton best for the early perods and gettng worse as the termnal perod approaches. We 7

8 shall see that ths rato s the exact result for the nfnte horzon verson of the problem. The followng Excel fle models ths problem, producng the optmal prce sequence, the dscounted and undscounted sum of profts, and the sequence {γ s }, whle allowng for any value of β [0, 1] and T {1,..., 95}: bc.edu/samson-alva/ec70f11/durablemonopoly.xls. () In general, we would need to keep track of the entre hstory of prces, but snce we know that today s demand s determned by the lowest prce set n the past, and that ths prce s the most recent prce set, as argued above, the only thng that needs to be kept track of s the last prce. Thus, the Bellman s V (p t 1 ) = max p t (p t (p t 1 p t ) + βv (p t )) (j) We have that the optmal prce p t = γp t 1 for any t > 0, and also that V (p t ) = A(p t ), where γ and A are coeffcents to be determned. The frst-order condton for optmalty s p t 1 p t + βv (p t ) = 0. But wth V (p t ) = A(p t ), so V (p t ) = Ap t. Usng ths, and the relatonshp p t = γp t 1, we obtan p t 1 = (1 βa)p t = 1 = (1 βa)γ. Next, the envelope theorem appled to the Bellman equaton mples V (p t 1 ) = p t. Usng our conjectures about both V and p t, we see that Ap t 1 = γp t 1, mplyng γ = A, = 4A(1 βa) = 1. Thus, combnng the two equatons nvolvng γ and A, we get γ( βγ) = 1 = βγ γ + 1 = 0. Solvng ths quadratc for γ gves us two roots, γ = 1 β (1 ± 1 β) of whch only the negatve root yelds a value between 0 and 1. Thus, we have that γ = 1 (1 1 β) and A = γ = 1 (1 1 β). Thus, p β β t = 1 (1 1 β)p β t 1, whch s the same prcng polcy obtaned as T approached nfnty n the fntehorzon verson of the problem. The dscounted sum of profts s just the value functon at tme 1, wth state varable havng value 1, and so s equal to 1 (1 1 β). Next, note that β 8

9 p t = γ t p 0 = γ t. So, the undscounted sum of profts s p t (p t 1 p t ) = t=1 γ t (γ t 1 γ t ) = (1 γ) t=1 t=1 γ t 1 = 1 γ γ (γ + γ 4 + γ ) = 1 γ γ γ 1 γ = γ 1 + γ = 1 1 β β β = 1 β 3 + β (k) We have to use L Hôptal s Rule to evaluaton the lmt of γ as β approaches 0. So, β 1 β lm = lm = 1 β 0 β β 0 1. Usng the expressons above for dscounted proft and undscounted proft, we see that wth β = 0, they are 1 and 1, respectvely. Essentally, γ = 1 s obtaned 4 3 because the monopolst doesn t care at all about the future, and so s really just solvng a sequence of statc optmzaton problems. Gven the lnear demand wth slope 1, the statcally optmal thng to do s to charge half of the maxmum valuaton n the market, and that s exactly what the monopolst does n every perod. The dscounted value of profts s just the proft n the current perod, snce the future doesn t matter at all. But the undscounted sum of profts s greater, though not as great as the total market surplus, whch equals the total area under the demand curve.e. 1. (l) The lmt of γ as β approaches 1 s 1. Thus, the optmal prcng rule s p t = p t 1. But ths doesn t make any sense, snce t mples that prce stays at 1. Nevertheless, the dscounted and undscounted profts (whch are the same n the lmtng case of β = 1) converge to 1, whch mples that the monopolst extracts all consumer surplus. So, magne a value of β very close but not equal to 1. The prce falls extremely slowly, and sales n each perod s very low. But, the consumers who buy n a partcular perod pay very close to ther valuaton, and so very lttle consumer surplus s generated. Instead, the monopolst succeeds n extractng almost all the surplus as profts. As the monopolst becomes more patent, t s more wllng to wat to earn these extra profts by very gradually decreasng prce. Notce that no matter what the dscount factor of the monopolst s, eventually every consumer wth non-zero valuaton wll get to buy the product, because the prce wll eventually fall below any postve number as t heads for 0. Thus, there s no deadweght loss n ths sense. However, f consumers dscount future utlty, clearly the delay nduces a cost, and so ths smplstc welfare analyss does not gve us the entre pcture. The model however does a good job of explanng the prcng polcy of technol- 9

10 ogy companes lke Apple wth Phones or Mcrosoft wth the Xbox, who greatly restrct the quantty of a new durable good ntally n order to rase the prce and extract the surplus from hgh-value consumers, and subsequently release more and more goods at lower prces. The rate of decrease of the prce depends on how mpatent the monopolst s. The lmtaton of ths model s that consumers are assumed to be myopc. In fact, people strategcally tme when to buy the durable, antcpatng that prces wll fall, and so a more robust analyss wll model ths stuaton as a game. 4. Optmal Allocaton wth Envous Consumers There s a fxed total Y > 0 of goods at the dsposal of socety. There are two consumers who envy each other. If consumer 1 gets y and consumer gets z, ther utltes are U 1 = y kz and U = z ky, respectvely, where k > 0 s a constant. The allocaton must satsfy y + z Y, and must maxmze U 1 + U. Also, all allocatons must be non-negatve. (a) The planner s problem s max y,z (y kz + z ky ) subject to y + z Y, y 0 and z 0. Defne the Lagrangan L (y kz + z ky ) λ(y + z Y ) = (1 λ)(y + z) k(y + z ) + λy. The Karush-Kuhn-Tucker condtons are L y = 1 λ ky 0, y 0, yl y = 0 L z = 1 λ kz 0, z 0, zl z = 0 L λ = Y y z 0, λ 0, λl λ = 0 Suppose λ > 0. Then y + z = Y. Snce Y > 0, at least one of y or z s postve. Wthout loss of generalty, suppose y > 0. Then, 1 λ ky = 0, whch mples λ = 1 ky. Then, 1 (1 ky) kz 0, whch mples k(y z) 0. But then y z 0, mplyng z y > 0, and thus we conclude that z > 0, and that L z = 0. So, we obtan λ = 1 kz, and thus y = z = Y. Also, λ = 1 ky. Suppose λ = 0. Then, y 1 > 0 and x 1 > 0. But complementary slackness k k condtons yl y = 0 and zl z = 0 then mply that L y = L z = 0, and so y = z = 1. k Thus, y + z = 1 Y. k (b) Suppose Y > 1. Now, let s suppose that the resource constrant bnds at the k optmum. Then t must be the case that λ > 0; n fact, we showed above that λ = 1 ky. But wth Y > 1 we get λ < 0, whch volates the KKT condtons. k Thus, the resource constrant cannot bnd the optmum n ths case. (c) The planner treats the two agents symmetrcally, but there are consumpton externaltes due to envy. The planner eventually encounters the stuaton that the margnal ncrease n envy n the socety exceeds the margnal drect beneft to the consumers of more goods. Thus, these goods have no value to agents or to the planner. 10

11 (d) The planner s problem s max {x } x k(x + x 3) subject to x Y, x 0 for all {1,, 3}. Defne the Lagrangan L x k(x + x 3) λ( x Y ) = (1 λ) x k(x + x 3) + λy. The Karush-Kuhn-Tucker condtons are {, 3}, L = 1 λ kx 0, x 0, x L = 0 L 1 = 1 λ 0, x 1 0, x 1 L 1 = 0 L λ = Y x 0, λ 0, λl λ = 0 Suppose x 1 > 0. Then λ = 1, whch mples that x = x 3 = 0, and snce the constrant s bndng, we get that x 1 = Y. Thus, (x 1, x, x 3, λ) = (Y, 0, 0, 1) solves the frst-order condtons. Next, suppose x 1 = 0. Then λ 1, whch mples that x 0, for {, 3}. But ths mples that x = 0, and so x =. Thus, the constrant s slack, whch mples that λ = 0. But ths yelds a contradcton wth the condton that λ 1. Thus, the only soluton s that the planner allocated all the resources to agent 1. The ntuton s that the consumpton externalty created by the envousness of agents nduces the planner to reduce the amount of envy created. But snce utlty s lnear for all agents, other than the envy component, the planner faces no dffculty n reducng the allocaton to agents that are enved and reallocatng the good to the unenved agent. (e) The ntuton from the prevous part contnues to hold. Agent 1 s the only agent that s not enved, and reducng the allocaton of some enved agent, say k, and correspondngly ncreasng the allocaton to agent 1 elmnates some envy, but the ncrease n agent 1 s utlty equal exactly the decrease n agent k s utlty, and so the planner acheves a hgher value of total utlty. Formally, the planner s problem s max {x } x k n = x subject to x Y, 11

12 x 0 for all {1,..., n}. Defne the Lagrangan L x k n x λ( = x Y ) = (1 λ) x k n x + λy. = The Karush-Kuhn-Tucker condtons are {,..., n}, L = 1 λ kx 0, x 0, x L = 0 L 1 = 1 λ 0, x 1 0, x 1 L 1 = 0 L λ = Y x 0, λ 0, λl λ = 0 Suppose x 1 > 0. Then λ = 1, whch mples that x = 0 for all {,..., n}, and snce the constrant s bndng, we get that x 1 = Y. Next, suppose x 1 = 0. Then λ 1, whch mples that x 0, for {,..., n}. But ths mples that x = 0, and so x =. Thus, the constrant s slack, whch mples that λ = 0. But ths yelds a contradcton wth the condton that λ 1. Thus, the only soluton s for the planner to allocate all of Y to agent 1. (f) The planner s problem s subject to max {x } (x kx ) x Y, x 0 for all {1,, 3}. Defne the Lagrangan L (x kx ) λ( x Y ) = (1 λ) x k x + λy. The Karush-Kuhn-Tucker condtons are L = 1 λ kx 0, x 0, x L = 0 L λ = Y x 0, λ 0, λl λ = 0 Suppose λ > 0. Then x = Y. Snce Y > 0, x > 0 for some ; w.l.o.g. suppose x 1 > 0. Then, λ = 1 kx 1, and so 1 (1 kx 1 ) kx 0, mplyng k(x 1 x ) 0. But then x 1 x 0 and so x x 1 > 0, so we conclude that x > 0 for all, and thus λ = 1 kx. Summng over all agents, 3λ = 3 ky, and so λ = 1 ky. Also, λ = 1 kx 3 = 1 kx j, so x = x j, and thus x = 1Y 3 for all. Suppose λ = 0. Then x 1 > 0 for all. But complementary slackness k condtons x L = 0 mples that L = 0 and thus x = 1 for all. Thus, ths k soluton obtans when x = 3 < Y. k The ntuton s smlar to that of part a). Snce all agents are enved (and symmetrcally), the planner wll dstrbute resources evenly. However, whle ntrnsc 1

13 utlty from the good s lnear, the dsutlty generated va envy s quadratc, and so eventually the planner wll stop dstrbutng the good because envousness ncreases at the margn more than the utlty of the recpent of the margnal unt of the good. Thus, t s possble that the resource constrant does not bnd. (g) Followng the argument of the prevous part, t s straghtforward to obtan the followng: f Y n then x k = Y for all ; f Y > n, then x n k = 1 for all. k Okay, now let s push ths generalzaton further. Suppose that you have an economy wth n agents, and agent has utlty functon U = x k n j=1 A jx j, where A j [0, 1] represents the degree of envy agent has for agent j s allocaton, where k > 0 captures the maxmum possble degree of envy. Assume that for all, A = 0; an agent doesn t envy hmself. Assume the planner wshes to maxmze the sum of utltes of all agents, n =1 U by allocaton the quantty of goods Y, subject to the constrants n =1 x Y, and x 0. (h) The Lagrangan s L = x k A j x j λ( j x Y ). Defne α j = A j; ths s a metrc of the degree to whch agent j s enved by other agents. The Lagrangan can be rewrtten as L = (1 λ) x k α x + λy. The KKT condtons are L = 1 λ kα x 0, x 0, x L = 0 L λ = ( x Y ) 0, λ 0, λl λ = 0 Note that α x 1 λ, and that equalty obtans f x k > 0, from the complementary slackness condton x L = 0. Let I be the set of agents that receve a strctly postve amount of the good at the optmum, and J be the set of agents that receve a zero allocaton at the optmum. For any I, α x = 1 λ, and α k x = α x for all, I. Notce that f α = 0 for some I, then α x = 0 for all I, and so α = 0. Thus, ether α > 0 for all I or α = 0 for all I. In words, we have the result that f any unenved agent (an agent wth α = 0) receves a postve allocaton, then all agents that receve a postve allocaton must be unenved. Also, we have the result that f any enved agent (an agent wth α > 0) receves a postve allocaton, then any agent that receves a postve allocaton must be enved. Now, consder an agent j J, assumng ths set s nonempty. Suppose there exsts some agent I. Snce α j x j 1 λ = α k x, we have that α j x j α x. But 13

14 snce j J, x j = 0, so α x 0, whch mples that α = 0, snce x > 0. In words, we have the result that f any agent receves a zero allocaton then only unenved agents can receve a strctly postve allocaton. Fnally, notce that f there exsts a j J and an I such that α j > 0 and α > 0, then we have that α x = 1 λ α k j x j, whch mples that x α j α x j. But j J mples that x j = 0, whch mples that x 0, whch contradcts our assumpton that I. Thus, n words, f any enved agent receves a strctly postve allocaton, then all enved agents receve a strctly postve allocaton. Puttng all the results together, we have the followng pcture. If there s at least one unenved agent, then every enved agent receves a zero allocaton, and at least one of the unenved agents receves a strctly postve allocaton. If there are no unenved agents, then every (enved) agent receves a strctly postve allocaton, snce Y > 0. Frst, let s suppose there are no unenved agents.e. α > 0 for all. Defne α H 1 α, the harmonc mean of the set {α }. Then, we know that x > 0 for all, and so α x = 1 λ. Consder two cases: λ > 0 mples k x = Y, so n(1 λ) = Y, so x kα H = αh Y. Next, f λ = 0, then x α n = 1 1 and so α k x = 1 n. α H k Now, let s suppose there s a set of unenved agents, denoted I. We know that only unenved agents can receve a strctly postve allocaton. Take some such agent. Then, α x = 1 λ. But snce α k = 0, ths mples that λ = 1, and so I x = Y. Thus, for any agent j, α j x j 0, and so we have a whole class of solutons defned as follows: every enved agent receves none of the good (x j = 0 for all j I), and any allocaton of the good that satsfed I x = Y and x 0 for all I s an allocaton. Ths set of solutons resembles an I -dmensonal smplex. () Suppose you are agent and the agent whom you envy more s agent j. Suppose agent j was unenved. If there s some other agent that s also unenved (possbly you), then your ncreased envy of agent j means hs new allocaton s zero, snce the presence of an unenved agent means that enved agents receve nothng, and so agent j s allocaton decreases (possbly strctly, snce an unenved agent can receve zero or a postve amount of the good). If you are an unenved agent, then there exst optmal solutons that result n you recevng a larger allocaton. If you are an enved agent, then your allocaton remans at zero. Lastly, suppose that agent j was the only unenved agent before your envy for hm ncreased. Then, he used to receve the entre amount Y, but now that he s enved, he wll receve strctly less than Y, and so your envy has decreased hs allocaton. Moreover, your allocaton strctly ncreases snce you used to receve zero, and now you ll receve some postve amount. Suppose agent j was enved. If there s some other agent that s unenved (possbly you), then your ncreased envy of j does nothng to hs allocaton, snce t s already zero. If you are also enved, then your allocaton also remans at zero. However, f you were unenved, there exst optmal allocatons where you wll receve strctly more now that your envy of j has ncreased. Lastly, suppose 14

15 there are no unenved agents. Then your ncreased envy of j mples that α j strctly ncreases, whch mples that x j strctly decreases. To see ths, notce that αh n α = 1+ α, whch clearly decreases when α j ncreases. There are two j α j cases to consder when examnng the effect of your ncreased envy of j on your own allocaton. Frst, suppose x > Y. Snce α H ncreases when α j ncreases, x = αh Y must strctly ncrease. Next, suppose that α n x Y. Now, the constrant s almost slack, and so x = 1 α, whch s unchanged. k To summarze, f agent s envy for agent j ncreases, then the planner s allocaton to agent j s weakly decreased and the allocaton to agent s weakly ncreased, and there exst stuatons where ths weak changes are strct. 15

k t+1 + c t A t k t, t=0

k t+1 + c t A t k t, t=0 Macro II (UC3M, MA/PhD Econ) Professor: Matthas Kredler Fnal Exam 6 May 208 You have 50 mnutes to complete the exam There are 80 ponts n total The exam has 4 pages If somethng n the queston s unclear,

More information

Economics 101. Lecture 4 - Equilibrium and Efficiency

Economics 101. Lecture 4 - Equilibrium and Efficiency Economcs 0 Lecture 4 - Equlbrum and Effcency Intro As dscussed n the prevous lecture, we wll now move from an envronment where we looed at consumers mang decsons n solaton to analyzng economes full of

More information

Module 9. Lecture 6. Duality in Assignment Problems

Module 9. Lecture 6. Duality in Assignment Problems Module 9 1 Lecture 6 Dualty n Assgnment Problems In ths lecture we attempt to answer few other mportant questons posed n earler lecture for (AP) and see how some of them can be explaned through the concept

More information

Equilibrium with Complete Markets. Instructor: Dmytro Hryshko

Equilibrium with Complete Markets. Instructor: Dmytro Hryshko Equlbrum wth Complete Markets Instructor: Dmytro Hryshko 1 / 33 Readngs Ljungqvst and Sargent. Recursve Macroeconomc Theory. MIT Press. Chapter 8. 2 / 33 Equlbrum n pure exchange, nfnte horzon economes,

More information

Supporting Information for: Two Monetary Models with Alternating Markets

Supporting Information for: Two Monetary Models with Alternating Markets Supportng Informaton for: Two Monetary Models wth Alternatng Markets Gabrele Camera Chapman Unversty & Unversty of Basel YL Chen St. Lous Fed November 2015 1 Optmal choces n the CIA model On date t, gven

More information

Supporting Materials for: Two Monetary Models with Alternating Markets

Supporting Materials for: Two Monetary Models with Alternating Markets Supportng Materals for: Two Monetary Models wth Alternatng Markets Gabrele Camera Chapman Unversty Unversty of Basel YL Chen Federal Reserve Bank of St. Lous 1 Optmal choces n the CIA model On date t,

More information

PROBLEM SET 7 GENERAL EQUILIBRIUM

PROBLEM SET 7 GENERAL EQUILIBRIUM PROBLEM SET 7 GENERAL EQUILIBRIUM Queston a Defnton: An Arrow-Debreu Compettve Equlbrum s a vector of prces {p t } and allocatons {c t, c 2 t } whch satsfes ( Gven {p t }, c t maxmzes βt ln c t subject

More information

Economics 2450A: Public Economics Section 10: Education Policies and Simpler Theory of Capital Taxation

Economics 2450A: Public Economics Section 10: Education Policies and Simpler Theory of Capital Taxation Economcs 2450A: Publc Economcs Secton 10: Educaton Polces and Smpler Theory of Captal Taxaton Matteo Parads November 14, 2016 In ths secton we study educaton polces n a smplfed verson of framework analyzed

More information

Welfare Properties of General Equilibrium. What can be said about optimality properties of resource allocation implied by general equilibrium?

Welfare Properties of General Equilibrium. What can be said about optimality properties of resource allocation implied by general equilibrium? APPLIED WELFARE ECONOMICS AND POLICY ANALYSIS Welfare Propertes of General Equlbrum What can be sad about optmalty propertes of resource allocaton mpled by general equlbrum? Any crteron used to compare

More information

Assortment Optimization under MNL

Assortment Optimization under MNL Assortment Optmzaton under MNL Haotan Song Aprl 30, 2017 1 Introducton The assortment optmzaton problem ams to fnd the revenue-maxmzng assortment of products to offer when the prces of products are fxed.

More information

Problem Set 9 Solutions

Problem Set 9 Solutions Desgn and Analyss of Algorthms May 4, 2015 Massachusetts Insttute of Technology 6.046J/18.410J Profs. Erk Demane, Srn Devadas, and Nancy Lynch Problem Set 9 Solutons Problem Set 9 Solutons Ths problem

More information

Notes on Kehoe Perri, Econometrica 2002

Notes on Kehoe Perri, Econometrica 2002 Notes on Kehoe Perr, Econometrca 2002 Jonathan Heathcote November 2nd 2005 There s nothng n these notes that s not n Kehoe Perr NBER Workng Paper 7820 or Kehoe and Perr Econometrca 2002. However, I have

More information

Online Appendix. t=1 (p t w)q t. Then the first order condition shows that

Online Appendix. t=1 (p t w)q t. Then the first order condition shows that Artcle forthcomng to ; manuscrpt no (Please, provde the manuscrpt number!) 1 Onlne Appendx Appendx E: Proofs Proof of Proposton 1 Frst we derve the equlbrum when the manufacturer does not vertcally ntegrate

More information

How Strong Are Weak Patents? Joseph Farrell and Carl Shapiro. Supplementary Material Licensing Probabilistic Patents to Cournot Oligopolists *

How Strong Are Weak Patents? Joseph Farrell and Carl Shapiro. Supplementary Material Licensing Probabilistic Patents to Cournot Oligopolists * How Strong Are Weak Patents? Joseph Farrell and Carl Shapro Supplementary Materal Lcensng Probablstc Patents to Cournot Olgopolsts * September 007 We study here the specal case n whch downstream competton

More information

Lecture 10 Support Vector Machines II

Lecture 10 Support Vector Machines II Lecture 10 Support Vector Machnes II 22 February 2016 Taylor B. Arnold Yale Statstcs STAT 365/665 1/28 Notes: Problem 3 s posted and due ths upcomng Frday There was an early bug n the fake-test data; fxed

More information

Additional Codes using Finite Difference Method. 1 HJB Equation for Consumption-Saving Problem Without Uncertainty

Additional Codes using Finite Difference Method. 1 HJB Equation for Consumption-Saving Problem Without Uncertainty Addtonal Codes usng Fnte Dfference Method Benamn Moll 1 HJB Equaton for Consumpton-Savng Problem Wthout Uncertanty Before consderng the case wth stochastc ncome n http://www.prnceton.edu/~moll/ HACTproect/HACT_Numercal_Appendx.pdf,

More information

Limited Dependent Variables

Limited Dependent Variables Lmted Dependent Varables. What f the left-hand sde varable s not a contnuous thng spread from mnus nfnty to plus nfnty? That s, gven a model = f (, β, ε, where a. s bounded below at zero, such as wages

More information

More metrics on cartesian products

More metrics on cartesian products More metrcs on cartesan products If (X, d ) are metrc spaces for 1 n, then n Secton II4 of the lecture notes we defned three metrcs on X whose underlyng topologes are the product topology The purpose of

More information

Lecture 14: Bandits with Budget Constraints

Lecture 14: Bandits with Budget Constraints IEOR 8100-001: Learnng and Optmzaton for Sequental Decson Makng 03/07/16 Lecture 14: andts wth udget Constrants Instructor: Shpra Agrawal Scrbed by: Zhpeng Lu 1 Problem defnton In the regular Mult-armed

More information

Lecture 21: Numerical methods for pricing American type derivatives

Lecture 21: Numerical methods for pricing American type derivatives Lecture 21: Numercal methods for prcng Amercan type dervatves Xaoguang Wang STAT 598W Aprl 10th, 2014 (STAT 598W) Lecture 21 1 / 26 Outlne 1 Fnte Dfference Method Explct Method Penalty Method (STAT 598W)

More information

Perfect Competition and the Nash Bargaining Solution

Perfect Competition and the Nash Bargaining Solution Perfect Competton and the Nash Barganng Soluton Renhard John Department of Economcs Unversty of Bonn Adenauerallee 24-42 53113 Bonn, Germany emal: rohn@un-bonn.de May 2005 Abstract For a lnear exchange

More information

University of California, Davis Date: June 22, 2009 Department of Agricultural and Resource Economics. PRELIMINARY EXAMINATION FOR THE Ph.D.

University of California, Davis Date: June 22, 2009 Department of Agricultural and Resource Economics. PRELIMINARY EXAMINATION FOR THE Ph.D. Unversty of Calforna, Davs Date: June 22, 29 Department of Agrcultural and Resource Economcs Department of Economcs Tme: 5 hours Mcroeconomcs Readng Tme: 2 mnutes PRELIMIARY EXAMIATIO FOR THE Ph.D. DEGREE

More information

Problem Set 4: Sketch of Solutions

Problem Set 4: Sketch of Solutions Problem Set 4: Sketc of Solutons Informaton Economcs (Ec 55) George Georgads Due n class or by e-mal to quel@bu.edu at :30, Monday, December 8 Problem. Screenng A monopolst can produce a good n dfferent

More information

princeton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg

princeton univ. F 17 cos 521: Advanced Algorithm Design Lecture 7: LP Duality Lecturer: Matt Weinberg prnceton unv. F 17 cos 521: Advanced Algorthm Desgn Lecture 7: LP Dualty Lecturer: Matt Wenberg Scrbe: LP Dualty s an extremely useful tool for analyzng structural propertes of lnear programs. Whle there

More information

Pricing and Resource Allocation Game Theoretic Models

Pricing and Resource Allocation Game Theoretic Models Prcng and Resource Allocaton Game Theoretc Models Zhy Huang Changbn Lu Q Zhang Computer and Informaton Scence December 8, 2009 Z. Huang, C. Lu, and Q. Zhang (CIS) Game Theoretc Models December 8, 2009

More information

CS286r Assign One. Answer Key

CS286r Assign One. Answer Key CS286r Assgn One Answer Key 1 Game theory 1.1 1.1.1 Let off-equlbrum strateges also be that people contnue to play n Nash equlbrum. Devatng from any Nash equlbrum s a weakly domnated strategy. That s,

More information

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009 College of Computer & Informaton Scence Fall 2009 Northeastern Unversty 20 October 2009 CS7880: Algorthmc Power Tools Scrbe: Jan Wen and Laura Poplawsk Lecture Outlne: Prmal-dual schema Network Desgn:

More information

(1 ) (1 ) 0 (1 ) (1 ) 0

(1 ) (1 ) 0 (1 ) (1 ) 0 Appendx A Appendx A contans proofs for resubmsson "Contractng Informaton Securty n the Presence of Double oral Hazard" Proof of Lemma 1: Assume that, to the contrary, BS efforts are achevable under a blateral

More information

Hila Etzion. Min-Seok Pang

Hila Etzion. Min-Seok Pang RESERCH RTICLE COPLEENTRY ONLINE SERVICES IN COPETITIVE RKETS: INTINING PROFITILITY IN THE PRESENCE OF NETWORK EFFECTS Hla Etzon Department of Technology and Operatons, Stephen. Ross School of usness,

More information

Lecture 4. Instructor: Haipeng Luo

Lecture 4. Instructor: Haipeng Luo Lecture 4 Instructor: Hapeng Luo In the followng lectures, we focus on the expert problem and study more adaptve algorthms. Although Hedge s proven to be worst-case optmal, one may wonder how well t would

More information

( ) 2 ( ) ( ) Problem Set 4 Suggested Solutions. Problem 1

( ) 2 ( ) ( ) Problem Set 4 Suggested Solutions. Problem 1 Problem Set 4 Suggested Solutons Problem (A) The market demand functon s the soluton to the followng utlty-maxmzaton roblem (UMP): The Lagrangean: ( x, x, x ) = + max U x, x, x x x x st.. x + x + x y x,

More information

Winter 2008 CS567 Stochastic Linear/Integer Programming Guest Lecturer: Xu, Huan

Winter 2008 CS567 Stochastic Linear/Integer Programming Guest Lecturer: Xu, Huan Wnter 2008 CS567 Stochastc Lnear/Integer Programmng Guest Lecturer: Xu, Huan Class 2: More Modelng Examples 1 Capacty Expanson Capacty expanson models optmal choces of the tmng and levels of nvestments

More information

Credit Card Pricing and Impact of Adverse Selection

Credit Card Pricing and Impact of Adverse Selection Credt Card Prcng and Impact of Adverse Selecton Bo Huang and Lyn C. Thomas Unversty of Southampton Contents Background Aucton model of credt card solctaton - Errors n probablty of beng Good - Errors n

More information

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X

3.1 Expectation of Functions of Several Random Variables. )' be a k-dimensional discrete or continuous random vector, with joint PMF p (, E X E X1 E X Statstcs 1: Probablty Theory II 37 3 EPECTATION OF SEVERAL RANDOM VARIABLES As n Probablty Theory I, the nterest n most stuatons les not on the actual dstrbuton of a random vector, but rather on a number

More information

Difference Equations

Difference Equations Dfference Equatons c Jan Vrbk 1 Bascs Suppose a sequence of numbers, say a 0,a 1,a,a 3,... s defned by a certan general relatonshp between, say, three consecutve values of the sequence, e.g. a + +3a +1

More information

MMA and GCMMA two methods for nonlinear optimization

MMA and GCMMA two methods for nonlinear optimization MMA and GCMMA two methods for nonlnear optmzaton Krster Svanberg Optmzaton and Systems Theory, KTH, Stockholm, Sweden. krlle@math.kth.se Ths note descrbes the algorthms used n the author s 2007 mplementatons

More information

Mixed Taxation and Production Efficiency

Mixed Taxation and Production Efficiency Floran Scheuer 2/23/2016 Mxed Taxaton and Producton Effcency 1 Overvew 1. Unform commodty taxaton under non-lnear ncome taxaton Atknson-Stgltz (JPubE 1976) Theorem Applcaton to captal taxaton 2. Unform

More information

ECE559VV Project Report

ECE559VV Project Report ECE559VV Project Report (Supplementary Notes Loc Xuan Bu I. MAX SUM-RATE SCHEDULING: THE UPLINK CASE We have seen (n the presentaton that, for downlnk (broadcast channels, the strategy maxmzng the sum-rate

More information

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix

Lectures - Week 4 Matrix norms, Conditioning, Vector Spaces, Linear Independence, Spanning sets and Basis, Null space and Range of a Matrix Lectures - Week 4 Matrx norms, Condtonng, Vector Spaces, Lnear Independence, Spannng sets and Bass, Null space and Range of a Matrx Matrx Norms Now we turn to assocatng a number to each matrx. We could

More information

Market structure and Innovation

Market structure and Innovation Market structure and Innovaton Ths presentaton s based on the paper Market structure and Innovaton authored by Glenn C. Loury, publshed n The Quarterly Journal of Economcs, Vol. 93, No.3 ( Aug 1979) I.

More information

f(x,y) = (4(x 2 4)x,2y) = 0 H(x,y) =

f(x,y) = (4(x 2 4)x,2y) = 0 H(x,y) = Problem Set 3: Unconstraned mzaton n R N. () Fnd all crtcal ponts of f(x,y) (x 4) +y and show whch are ma and whch are mnma. () Fnd all crtcal ponts of f(x,y) (y x ) x and show whch are ma and whch are

More information

Week3, Chapter 4. Position and Displacement. Motion in Two Dimensions. Instantaneous Velocity. Average Velocity

Week3, Chapter 4. Position and Displacement. Motion in Two Dimensions. Instantaneous Velocity. Average Velocity Week3, Chapter 4 Moton n Two Dmensons Lecture Quz A partcle confned to moton along the x axs moves wth constant acceleraton from x =.0 m to x = 8.0 m durng a 1-s tme nterval. The velocty of the partcle

More information

Lecture Notes on Linear Regression

Lecture Notes on Linear Regression Lecture Notes on Lnear Regresson Feng L fl@sdueducn Shandong Unversty, Chna Lnear Regresson Problem In regresson problem, we am at predct a contnuous target value gven an nput feature vector We assume

More information

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module 3 LOSSY IMAGE COMPRESSION SYSTEMS Verson ECE IIT, Kharagpur Lesson 6 Theory of Quantzaton Verson ECE IIT, Kharagpur Instructonal Objectves At the end of ths lesson, the students should be able to:

More information

Solutions HW #2. minimize. Ax = b. Give the dual problem, and make the implicit equality constraints explicit. Solution.

Solutions HW #2. minimize. Ax = b. Give the dual problem, and make the implicit equality constraints explicit. Solution. Solutons HW #2 Dual of general LP. Fnd the dual functon of the LP mnmze subject to c T x Gx h Ax = b. Gve the dual problem, and make the mplct equalty constrants explct. Soluton. 1. The Lagrangan s L(x,

More information

The Second Anti-Mathima on Game Theory

The Second Anti-Mathima on Game Theory The Second Ant-Mathma on Game Theory Ath. Kehagas December 1 2006 1 Introducton In ths note we wll examne the noton of game equlbrum for three types of games 1. 2-player 2-acton zero-sum games 2. 2-player

More information

In the figure below, the point d indicates the location of the consumer that is under competition. Transportation costs are given by td.

In the figure below, the point d indicates the location of the consumer that is under competition. Transportation costs are given by td. UC Berkeley Economcs 11 Sprng 006 Prof. Joseph Farrell / GSI: Jenny Shanefelter Problem Set # - Suggested Solutons. 1.. In ths problem, we are extendng the usual Hotellng lne so that now t runs from [-a,

More information

j) = 1 (note sigma notation) ii. Continuous random variable (e.g. Normal distribution) 1. density function: f ( x) 0 and f ( x) dx = 1

j) = 1 (note sigma notation) ii. Continuous random variable (e.g. Normal distribution) 1. density function: f ( x) 0 and f ( x) dx = 1 Random varables Measure of central tendences and varablty (means and varances) Jont densty functons and ndependence Measures of assocaton (covarance and correlaton) Interestng result Condtonal dstrbutons

More information

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification

2E Pattern Recognition Solutions to Introduction to Pattern Recognition, Chapter 2: Bayesian pattern classification E395 - Pattern Recognton Solutons to Introducton to Pattern Recognton, Chapter : Bayesan pattern classfcaton Preface Ths document s a soluton manual for selected exercses from Introducton to Pattern Recognton

More information

Lecture Notes, January 11, 2010

Lecture Notes, January 11, 2010 Economcs 200B UCSD Wnter 2010 Lecture otes, January 11, 2010 Partal equlbrum comparatve statcs Partal equlbrum: Market for one good only wth supply and demand as a functon of prce. Prce s defned as the

More information

CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE

CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE CHAPTER 5 NUMERICAL EVALUATION OF DYNAMIC RESPONSE Analytcal soluton s usually not possble when exctaton vares arbtrarly wth tme or f the system s nonlnear. Such problems can be solved by numercal tmesteppng

More information

Problem Set 9 - Solutions Due: April 27, 2005

Problem Set 9 - Solutions Due: April 27, 2005 Problem Set - Solutons Due: Aprl 27, 2005. (a) Frst note that spam messages, nvtatons and other e-mal are all ndependent Posson processes, at rates pλ, qλ, and ( p q)λ. The event of the tme T at whch you

More information

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4)

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4) I. Classcal Assumptons Econ7 Appled Econometrcs Topc 3: Classcal Model (Studenmund, Chapter 4) We have defned OLS and studed some algebrac propertes of OLS. In ths topc we wll study statstcal propertes

More information

Stanford University CS359G: Graph Partitioning and Expanders Handout 4 Luca Trevisan January 13, 2011

Stanford University CS359G: Graph Partitioning and Expanders Handout 4 Luca Trevisan January 13, 2011 Stanford Unversty CS359G: Graph Parttonng and Expanders Handout 4 Luca Trevsan January 3, 0 Lecture 4 In whch we prove the dffcult drecton of Cheeger s nequalty. As n the past lectures, consder an undrected

More information

Lecture 12: Discrete Laplacian

Lecture 12: Discrete Laplacian Lecture 12: Dscrete Laplacan Scrbe: Tanye Lu Our goal s to come up wth a dscrete verson of Laplacan operator for trangulated surfaces, so that we can use t n practce to solve related problems We are mostly

More information

A NOTE ON CES FUNCTIONS Drago Bergholt, BI Norwegian Business School 2011

A NOTE ON CES FUNCTIONS Drago Bergholt, BI Norwegian Business School 2011 A NOTE ON CES FUNCTIONS Drago Bergholt, BI Norwegan Busness School 2011 Functons featurng constant elastcty of substtuton CES are wdely used n appled economcs and fnance. In ths note, I do two thngs. Frst,

More information

CS294 Topics in Algorithmic Game Theory October 11, Lecture 7

CS294 Topics in Algorithmic Game Theory October 11, Lecture 7 CS294 Topcs n Algorthmc Game Theory October 11, 2011 Lecture 7 Lecturer: Chrstos Papadmtrou Scrbe: Wald Krchene, Vjay Kamble 1 Exchange economy We consder an exchange market wth m agents and n goods. Agent

More information

,, MRTS is the marginal rate of technical substitution

,, MRTS is the marginal rate of technical substitution Mscellaneous Notes on roducton Economcs ompled by eter F Orazem September 9, 00 I Implcatons of conve soquants Two nput case, along an soquant 0 along an soquant Slope of the soquant,, MRTS s the margnal

More information

Economics 8105 Macroeconomic Theory Recitation 1

Economics 8105 Macroeconomic Theory Recitation 1 Economcs 8105 Macroeconomc Theory Rectaton 1 Outlne: Conor Ryan September 6th, 2016 Adapted From Anh Thu (Monca) Tran Xuan s Notes Last Updated September 20th, 2016 Dynamc Economc Envronment Arrow-Debreu

More information

= z 20 z n. (k 20) + 4 z k = 4

= z 20 z n. (k 20) + 4 z k = 4 Problem Set #7 solutons 7.2.. (a Fnd the coeffcent of z k n (z + z 5 + z 6 + z 7 + 5, k 20. We use the known seres expanson ( n+l ( z l l z n below: (z + z 5 + z 6 + z 7 + 5 (z 5 ( + z + z 2 + z + 5 5

More information

3.2. Cournot Model Cournot Model

3.2. Cournot Model Cournot Model Matlde Machado Assumptons: All frms produce an homogenous product The market prce s therefore the result of the total supply (same prce for all frms) Frms decde smultaneously how much to produce Quantty

More information

Managing Capacity Through Reward Programs. on-line companion page. Byung-Do Kim Seoul National University College of Business Administration

Managing Capacity Through Reward Programs. on-line companion page. Byung-Do Kim Seoul National University College of Business Administration Managng Caacty Through eward Programs on-lne comanon age Byung-Do Km Seoul Natonal Unversty College of Busness Admnstraton Mengze Sh Unversty of Toronto otman School of Management Toronto ON M5S E6 Canada

More information

EEL 6266 Power System Operation and Control. Chapter 3 Economic Dispatch Using Dynamic Programming

EEL 6266 Power System Operation and Control. Chapter 3 Economic Dispatch Using Dynamic Programming EEL 6266 Power System Operaton and Control Chapter 3 Economc Dspatch Usng Dynamc Programmng Pecewse Lnear Cost Functons Common practce many utltes prefer to represent ther generator cost functons as sngle-

More information

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems Numercal Analyss by Dr. Anta Pal Assstant Professor Department of Mathematcs Natonal Insttute of Technology Durgapur Durgapur-713209 emal: anta.bue@gmal.com 1 . Chapter 5 Soluton of System of Lnear Equatons

More information

NUMERICAL DIFFERENTIATION

NUMERICAL DIFFERENTIATION NUMERICAL DIFFERENTIATION 1 Introducton Dfferentaton s a method to compute the rate at whch a dependent output y changes wth respect to the change n the ndependent nput x. Ths rate of change s called the

More information

APPENDIX A Some Linear Algebra

APPENDIX A Some Linear Algebra APPENDIX A Some Lnear Algebra The collecton of m, n matrces A.1 Matrces a 1,1,..., a 1,n A = a m,1,..., a m,n wth real elements a,j s denoted by R m,n. If n = 1 then A s called a column vector. Smlarly,

More information

Endogenous timing in a mixed oligopoly consisting of a single public firm and foreign competitors. Abstract

Endogenous timing in a mixed oligopoly consisting of a single public firm and foreign competitors. Abstract Endogenous tmng n a mxed olgopoly consstng o a sngle publc rm and oregn compettors Yuanzhu Lu Chna Economcs and Management Academy, Central Unversty o Fnance and Economcs Abstract We nvestgate endogenous

More information

PHYS 705: Classical Mechanics. Calculus of Variations II

PHYS 705: Classical Mechanics. Calculus of Variations II 1 PHYS 705: Classcal Mechancs Calculus of Varatons II 2 Calculus of Varatons: Generalzaton (no constrant yet) Suppose now that F depends on several dependent varables : We need to fnd such that has a statonary

More information

Maximizing the number of nonnegative subsets

Maximizing the number of nonnegative subsets Maxmzng the number of nonnegatve subsets Noga Alon Hao Huang December 1, 213 Abstract Gven a set of n real numbers, f the sum of elements of every subset of sze larger than k s negatve, what s the maxmum

More information

THE CHINESE REMAINDER THEOREM. We should thank the Chinese for their wonderful remainder theorem. Glenn Stevens

THE CHINESE REMAINDER THEOREM. We should thank the Chinese for their wonderful remainder theorem. Glenn Stevens THE CHINESE REMAINDER THEOREM KEITH CONRAD We should thank the Chnese for ther wonderful remander theorem. Glenn Stevens 1. Introducton The Chnese remander theorem says we can unquely solve any par of

More information

Unit 5: Government policy in competitive markets I E ciency

Unit 5: Government policy in competitive markets I E ciency Unt 5: Government polcy n compettve markets I E cency Prof. Antono Rangel January 2, 2016 1 Pareto optmal allocatons 1.1 Prelmnares Bg pcture Consumers: 1,...,C,eachw/U,W Frms: 1,...,F,eachw/C ( ) Consumers

More information

Online Appendix to The Allocation of Talent and U.S. Economic Growth

Online Appendix to The Allocation of Talent and U.S. Economic Growth Onlne Appendx to The Allocaton of Talent and U.S. Economc Growth Not for publcaton) Chang-Ta Hseh, Erk Hurst, Charles I. Jones, Peter J. Klenow February 22, 23 A Dervatons and Proofs The propostons n the

More information

Cournot Equilibrium with N firms

Cournot Equilibrium with N firms Recap Last class (September 8, Thursday) Examples of games wth contnuous acton sets Tragedy of the commons Duopoly models: ournot o class on Sept. 13 due to areer Far Today (September 15, Thursday) Duopoly

More information

Structure and Drive Paul A. Jensen Copyright July 20, 2003

Structure and Drive Paul A. Jensen Copyright July 20, 2003 Structure and Drve Paul A. Jensen Copyrght July 20, 2003 A system s made up of several operatons wth flow passng between them. The structure of the system descrbes the flow paths from nputs to outputs.

More information

Kernel Methods and SVMs Extension

Kernel Methods and SVMs Extension Kernel Methods and SVMs Extenson The purpose of ths document s to revew materal covered n Machne Learnng 1 Supervsed Learnng regardng support vector machnes (SVMs). Ths document also provdes a general

More information

Vickrey Auction VCG Combinatorial Auctions. Mechanism Design. Algorithms and Data Structures. Winter 2016

Vickrey Auction VCG Combinatorial Auctions. Mechanism Design. Algorithms and Data Structures. Winter 2016 Mechansm Desgn Algorthms and Data Structures Wnter 2016 1 / 39 Vckrey Aucton Vckrey-Clarke-Groves Mechansms Sngle-Mnded Combnatoral Auctons 2 / 39 Mechansm Desgn (wth Money) Set A of outcomes to choose

More information

ANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U)

ANSWERS. Problem 1. and the moment generating function (mgf) by. defined for any real t. Use this to show that E( U) var( U) Econ 413 Exam 13 H ANSWERS Settet er nndelt 9 deloppgaver, A,B,C, som alle anbefales å telle lkt for å gøre det ltt lettere å stå. Svar er gtt . Unfortunately, there s a prntng error n the hnt of

More information

Lecture 4 Hypothesis Testing

Lecture 4 Hypothesis Testing Lecture 4 Hypothess Testng We may wsh to test pror hypotheses about the coeffcents we estmate. We can use the estmates to test whether the data rejects our hypothess. An example mght be that we wsh to

More information

COS 521: Advanced Algorithms Game Theory and Linear Programming

COS 521: Advanced Algorithms Game Theory and Linear Programming COS 521: Advanced Algorthms Game Theory and Lnear Programmng Moses Charkar February 27, 2013 In these notes, we ntroduce some basc concepts n game theory and lnear programmng (LP). We show a connecton

More information

The optimal delay of the second test is therefore approximately 210 hours earlier than =2.

The optimal delay of the second test is therefore approximately 210 hours earlier than =2. THE IEC 61508 FORMULAS 223 The optmal delay of the second test s therefore approxmately 210 hours earler than =2. 8.4 The IEC 61508 Formulas IEC 61508-6 provdes approxmaton formulas for the PF for smple

More information

Errors for Linear Systems

Errors for Linear Systems Errors for Lnear Systems When we solve a lnear system Ax b we often do not know A and b exactly, but have only approxmatons  and ˆb avalable. Then the best thng we can do s to solve ˆx ˆb exactly whch

More information

The Multiple Classical Linear Regression Model (CLRM): Specification and Assumptions. 1. Introduction

The Multiple Classical Linear Regression Model (CLRM): Specification and Assumptions. 1. Introduction ECONOMICS 5* -- NOTE (Summary) ECON 5* -- NOTE The Multple Classcal Lnear Regresson Model (CLRM): Specfcaton and Assumptons. Introducton CLRM stands for the Classcal Lnear Regresson Model. The CLRM s also

More information

Math1110 (Spring 2009) Prelim 3 - Solutions

Math1110 (Spring 2009) Prelim 3 - Solutions Math 1110 (Sprng 2009) Solutons to Prelm 3 (04/21/2009) 1 Queston 1. (16 ponts) Short answer. Math1110 (Sprng 2009) Prelm 3 - Solutons x a 1 (a) (4 ponts) Please evaluate lm, where a and b are postve numbers.

More information

Price competition with capacity constraints. Consumers are rationed at the low-price firm. But who are the rationed ones?

Price competition with capacity constraints. Consumers are rationed at the low-price firm. But who are the rationed ones? Prce competton wth capacty constrants Consumers are ratoned at the low-prce frm. But who are the ratoned ones? As before: two frms; homogeneous goods. Effcent ratonng If p < p and q < D(p ), then the resdual

More information

Applied Stochastic Processes

Applied Stochastic Processes STAT455/855 Fall 23 Appled Stochastc Processes Fnal Exam, Bref Solutons 1. (15 marks) (a) (7 marks) The dstrbuton of Y s gven by ( ) ( ) y 2 1 5 P (Y y) for y 2, 3,... The above follows because each of

More information

APPROXIMATE PRICES OF BASKET AND ASIAN OPTIONS DUPONT OLIVIER. Premia 14

APPROXIMATE PRICES OF BASKET AND ASIAN OPTIONS DUPONT OLIVIER. Premia 14 APPROXIMAE PRICES OF BASKE AND ASIAN OPIONS DUPON OLIVIER Prema 14 Contents Introducton 1 1. Framewor 1 1.1. Baset optons 1.. Asan optons. Computng the prce 3. Lower bound 3.1. Closed formula for the prce

More information

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography CSc 6974 and ECSE 6966 Math. Tech. for Vson, Graphcs and Robotcs Lecture 21, Aprl 17, 2006 Estmatng A Plane Homography Overvew We contnue wth a dscusson of the major ssues, usng estmaton of plane projectve

More information

Econ674 Economics of Natural Resources and the Environment

Econ674 Economics of Natural Resources and the Environment Econ674 Economcs of Natural Resources and the Envronment Sesson 7 Exhaustble Resource Dynamc An Introducton to Exhaustble Resource Prcng 1. The dstncton between nonrenewable and renewable resources can

More information

The Geometry of Logit and Probit

The Geometry of Logit and Probit The Geometry of Logt and Probt Ths short note s meant as a supplement to Chapters and 3 of Spatal Models of Parlamentary Votng and the notaton and reference to fgures n the text below s to those two chapters.

More information

and problem sheet 2

and problem sheet 2 -8 and 5-5 problem sheet Solutons to the followng seven exercses and optonal bonus problem are to be submtted through gradescope by :0PM on Wednesday th September 08. There are also some practce problems,

More information

January Examinations 2015

January Examinations 2015 24/5 Canddates Only January Examnatons 25 DO NOT OPEN THE QUESTION PAPER UNTIL INSTRUCTED TO DO SO BY THE CHIEF INVIGILATOR STUDENT CANDIDATE NO.. Department Module Code Module Ttle Exam Duraton (n words)

More information

Norm Bounds for a Transformed Activity Level. Vector in Sraffian Systems: A Dual Exercise

Norm Bounds for a Transformed Activity Level. Vector in Sraffian Systems: A Dual Exercise ppled Mathematcal Scences, Vol. 4, 200, no. 60, 2955-296 Norm Bounds for a ransformed ctvty Level Vector n Sraffan Systems: Dual Exercse Nkolaos Rodousaks Department of Publc dmnstraton, Panteon Unversty

More information

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017

U.C. Berkeley CS294: Beyond Worst-Case Analysis Luca Trevisan September 5, 2017 U.C. Berkeley CS94: Beyond Worst-Case Analyss Handout 4s Luca Trevsan September 5, 07 Summary of Lecture 4 In whch we ntroduce semdefnte programmng and apply t to Max Cut. Semdefnte Programmng Recall that

More information

Introduction to Vapor/Liquid Equilibrium, part 2. Raoult s Law:

Introduction to Vapor/Liquid Equilibrium, part 2. Raoult s Law: CE304, Sprng 2004 Lecture 4 Introducton to Vapor/Lqud Equlbrum, part 2 Raoult s Law: The smplest model that allows us do VLE calculatons s obtaned when we assume that the vapor phase s an deal gas, and

More information

Games of Threats. Elon Kohlberg Abraham Neyman. Working Paper

Games of Threats. Elon Kohlberg Abraham Neyman. Working Paper Games of Threats Elon Kohlberg Abraham Neyman Workng Paper 18-023 Games of Threats Elon Kohlberg Harvard Busness School Abraham Neyman The Hebrew Unversty of Jerusalem Workng Paper 18-023 Copyrght 2017

More information

Suggested solutions for the exam in SF2863 Systems Engineering. June 12,

Suggested solutions for the exam in SF2863 Systems Engineering. June 12, Suggested solutons for the exam n SF2863 Systems Engneerng. June 12, 2012 14.00 19.00 Examner: Per Enqvst, phone: 790 62 98 1. We can thnk of the farm as a Jackson network. The strawberry feld s modelled

More information

Appendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis

Appendix for Causal Interaction in Factorial Experiments: Application to Conjoint Analysis A Appendx for Causal Interacton n Factoral Experments: Applcaton to Conjont Analyss Mathematcal Appendx: Proofs of Theorems A. Lemmas Below, we descrbe all the lemmas, whch are used to prove the man theorems

More information

Physics 5153 Classical Mechanics. Principle of Virtual Work-1

Physics 5153 Classical Mechanics. Principle of Virtual Work-1 P. Guterrez 1 Introducton Physcs 5153 Classcal Mechancs Prncple of Vrtual Work The frst varatonal prncple we encounter n mechancs s the prncple of vrtual work. It establshes the equlbrum condton of a mechancal

More information

THE ROYAL STATISTICAL SOCIETY 2006 EXAMINATIONS SOLUTIONS HIGHER CERTIFICATE

THE ROYAL STATISTICAL SOCIETY 2006 EXAMINATIONS SOLUTIONS HIGHER CERTIFICATE THE ROYAL STATISTICAL SOCIETY 6 EXAMINATIONS SOLUTIONS HIGHER CERTIFICATE PAPER I STATISTICAL THEORY The Socety provdes these solutons to assst canddates preparng for the eamnatons n future years and for

More information

Statistics for Economics & Business

Statistics for Economics & Business Statstcs for Economcs & Busness Smple Lnear Regresson Learnng Objectves In ths chapter, you learn: How to use regresson analyss to predct the value of a dependent varable based on an ndependent varable

More information