Hypothesis testing 2

Hypothesis testing 2 - Test to perform Sides Hypotheses...

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

View Full Document Right Arrow Icon
Test to perform Sides Hypotheses Confidence interval Test statistic P-value P-value in Stata Notes Two- sided H 0 : μ = * H A : * If * is outside X ± t α , n 1 s x n If X * s x n > t 2 , n 1 If 2 P T n 1 > X * s x n < “ttest X = * Or: “display 2*ttail( n 1 ,( X - * )/( s x / n ^0.5))” Or: “reg X ”; “test _cons= * If n is large, then z -distribution is approximately correct in the manual calculations. In Stata: use either X * or * X , whichever is positive. Mean One- sided H 0 : = * H A : < * or H 0 : > * H A : < * If * is outside X ± t , n 1 s x n If X * s x n > t , n 1 If P T n 1 > X * s x n < “ttest X = * Or: “display ttail( n -1,( X - * )/( s x / n ^0.5)) Or: “reg X ”; “test _cons= * If n is large, then z -distribution is approximately correct. Same test if the direction of the inequalities is switched. In Stata: use either X * or * X , whichever is positive. When using “regress” in Stata, divide the p-value in half for a one-tailed test. Two- sided H 0 : p = p * H A : p p * If p * is outside ˆ p ± z ˆ p (1 ˆ p ) n If ˆ p p * ˆ p (1 ˆ p )/ n > z If 2 P Z > ˆ p p * ˆ p (1 ˆ p )/ n <
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 01/11/2012 for the course ECON 400 taught by Professor Turchi during the Fall '08 term at UNC.

Ask a homework question - tutors are online