ps5_ans - Department of Economics University of California...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Department of Economics Fall 2007 University of California Prof. Woroch Economics 140: Problem Set 5 Answer Sheet Instructions: Include names of all of the members of your study group and your common official GSI. When using Excel, include the final answer and provide the relevant portion of the Excel printout. Include all intermediate steps and indicate clearly the final solution. Staple your answer sheets. 1. A recent paper 1 examined some economic determinants of terrorism. In particular, the authors ran a logit model that predicts the likelihood of “death in a Hezbollah military event” based on an individual’s economic and educational background as well as age and area of residency. The results of the maximum likelihood estimation are given in the table below. a) Compute the estimated effect of a secondary education on the probability of death for a 20 year old who falls below the poverty line and is a resident of Beirut in Southern Lebanon. Answer : This effect can be computed by evaluating the logistic expression under the two hypothetical situations which differ by whether or not the individual has a secondary education. Recall the logistic expression for the probability of a “positive” event: Pr{ Y =1} = 1/{1+ exp [–( β 0 + β 1 X 1 + β 2 X 2 + … )]}. We can then compare probabilities before and after a secondary education taking the coefficient estimates from Model (2): Pr{ Y =1 | secondary = 1, poverty = 1, age = 20, Beirut = 1, South Lebanon = 1) - Pr{ Y =0 | secondary = 0, poverty = 1, age = 20, Beirut = 1, South Lebanon = 1) = 1/[1+ e –(–5.910 + 1 × 0.171 + 1 × (-0.167) + 20 × (-0.083) + 2.199 + 2.187) ] - 1/[1+ e –(–5.910 + 1 × (-0.167) + 20 × (-0.083) + 2.199 + 2.187) ] = 1/[1 + e -(-3.18) ] - 1/[1 + e -(-3.351) ] = 0.03993 - 0.03381 = 0.00606 Thus, consistent with the positive logit coefficient on the secondary education dummy, this calculation shows that higher education is associated with greater chance of participating in Hezbollah . This may seem like a very small amount since it is about ½ of 1 percent, but note that, relative to a baseline probability of 0.03381 or about 3.4%, a secondary education results in an 18% increase in the probability. On the other hand, the coefficient on secondary education is not statistically significant since the t-stat is 0.886. This same calculation could be done using the coefficient estimates from Model (4) and both calculations are included in the Excel file accompanying this answer sheet. Note that in fact it could not be the case that, as stated in the question, an individual was both from Beirut and from South Lebanon. These two areas are mutually exclusive. We can still go through the calculation using the estimated coefficients on the two dummy variables, however. Given these dummies, the excluded group consists of residents living outside both Beirut and South Lebanon, who are above the poverty line, and who have less than a secondary education.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 10/21/2008 for the course ECON 140 taught by Professor Duncan during the Fall '08 term at Berkeley.

Page1 / 10

ps5_ans - Department of Economics University of California...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online