Econometrics: Answer all Questions Not Parts. Use the following table from a stu

Question

Econometrics: Answer all Questions Not Parts.

Use the following table from a study by Burde and Linden (2012) to answer each of the treatment villages) but not others (control villages) in a rural area of Afghanistan in order to evaluate the effect of increased accessibility to schools. The table reports mean attendance outcomes for children in the treatment villages and for children in the control villages, as well as the difference in the means. The standard error for each difference in means estimate is reported in parentheses below the estimate.

Explanation / Answer

(a)Start with the first row in the table. Use to denote the difference in population

means. The null hypothesis is H0 : = 0 so the test statistic should be

T =/SE() ˆ

. Both the numerator and denominator are given in the table.

The difference in sample means is = 0 ˆ .467. The standard error is 0.085.

So T = 5.5. This is larger than the critical value of 1.96 so we reject the

null. For the next 5 rows the t statistics are .381/.04 = 9.5, .091/.041 = 2.2,

.470/.084 = 5.6, .376/.04 = 9.4, and .089/.045 = 1.98. Since these are all

greater than 1.96 we reject each of the null hypotheses. Across the board

we can conclude that there is sufficient statistical evidence of an effect of the intervention.

(b) The formula for a 95% confidence interval is to take the estimate ± the standard

error times the critical value, which is c0.05 = 1.96 for a 95% confidence interval.

Therefore, the six confidence intervals are [.300, .634], [.303, .459], [.011, .171],

[.305, .635], [.298, .454] and [.001, .177].

Related Questions

Navigate

• Browse (All)
• Browse E
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.