This preview shows pages 1–5. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 139Spring 09 FINALName ..................................................................DO NOT TURN OVER UNTIL TOLD TO DO SONotes:1. Use a calculator where necessary.2. Show your work, place your answers in the spaces provided and make use of the cheat sheet.3. If you want to make any additional assumptions to simplify a problem, please state them clearly.4.Keep verbal answers concise(no more than 4 sentences, and bullet points are &ne).Guide:There are 126 points and you have 150 minutes.Q1 and Q2 are short.Q3 is on treatment e/ects and part (e) contains proofs.Q4 is on IV.The questions can be done in any order.Good Luck and Enjoy the Summer.11. [18 points] You are working as an intern for a local bank. As part of a costcutting exercise,they want to know how many minutes it takes a teller to serve a customer. This type of duration datais commonly modeled using an exponential distribution. The pdf for a duration of lengthyiisf(yij&) =1&e&yi&foryi&and& >where&is the single parameter.(a) [4 points] Suppose that&= 1. Sketch the pdf on the axes below, marking the values of thepdf fory= 0; y= 1andy= 2.f(y)y12f(y)y12(b) You want to estimate the parameter&using maximum likelihood and a random sample ofNobservations from local branches.(i) [4 points] What is the joint probability of the observed data given&?2(ii) [4 points] Show that the loglikelihood is&Nlog(&)&PNi=1yi&(iii) [4 points] Derive the ML estimator of&(don&t worry about the secondorder condition).(iv) [2 points]N= 200;PNi=1yi= 424:3;PNi=1y2i= 1985:8. Calculate the ML point estimate of&.32. [16 points] You get data on a sample of students graduating from high school, together withan indicator for whether they will attend college in the Fall. You estimate probit models for whetherthey will go to college, as a function of individual and highschool characteristics. There are two speci&cations: one with high school characteristics and one without. Standard errors are in parentheses.(1)(2)Dep VarGo to CollegeGo to CollegeParental Income:04:03($000s)(0:01)(0:01)Number of Siblings (count)&:42&:27(0:20)(0:14)At least One Parent:95:90Attended College (Dummy)(0:34)(0:31)Private High School:80(Dummy)(0:23)Religious High School&:34(Dummy)(0:15)Constant&2:80&2:75(0:95)(0:85)Loglikelihood&4;300:78&4;280:78(a) [4 points] Based on the speci&cation in column (1) what is the predicted probability of attendingcollege for someone with no siblings, a parental income of $100,000 and parents who attended college.(b) [6 points] Perform a likelihood ratio test for whether the school characteristics are jointlysigni&cant at the 1% level. Show your working (i.e., the test statistic, the distribution used and thecritical value) clearly....
View
Full
Document
This note was uploaded on 08/02/2011 for the course ECON 139 taught by Professor Alessandrotarozzi during the Spring '08 term at Duke.
 Spring '08
 ALESSANDROTAROZZI
 Econometrics

Click to edit the document details