TA11 - S p 2 = 4 and construct a 95 C.I for μ X – μ Y as ³ ´ − µ ¸ ¹ º ⁄(2 − 2 ¼ ½ ¾ 2 =(16 − 13 À Á ÂÃ Ä

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
ECO2121: Methods of Economic Statistics TA11 1 April 2009 1. Confidence Interval (Two Populations) (continued) Motivation: previously we can construct a C.I. for two independent samples, how about two dependent samples? Example: According to the budget plan, an increase in consumption tax on cigarettes can motivate smokers to change their habit, such as reduce their consumption. Taylor asks n = 18 smokers before and after the budget plan, denote X = amount of cigarette consumption a day before the budget plan, and Y after, respectively, suppose ̅ = 16, s X 2 = 4, and ±² = 13, s Y 2 = 4. Assume σ X 2 = σ Y 2 , Taylor pools the variance as S
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: S p 2 = 4, and construct a 95% C.I. for μ X – μ Y as ( ³ ´ − µ ¶ ) · ¸ ¹ º ⁄ (2 » − 2) ¼ ½ ¾ 2 ¿ = (16 − 13) · À Á . ÂÃ Ä ⁄ (34) × 2 × 1 3 What ’ s wrong? Instead, Taylor this time constructs a new set of data as D i = X i – Y i , and compute the sample mean and the sample variance of D , for instance, Å Æ = 1.2 and S D 2 = 0.64. Then he constructs the C.I. as ÇÈ É · Ê Ë Ì ⁄ ( Í − 1) × ÎÏ Ð √Ñ ⁄ ÒÓ . What is the difference? What assumption does he need? Compute the C.I. for this example. Taylor 23 March 2009...
View Full Document

This note was uploaded on 11/16/2009 for the course ECO 2121 taught by Professor Professorwen during the Fall '08 term at Al Ahliyya Amman University.

Ask a homework question - tutors are online