Econometrics-I-8

# Part 8 hypothesis testing methodology p bayesian n

This preview shows pages 12–19. Sign up to view the full content.

Part 8: Hypothesis Testing Methodology p Bayesian n Prior odds compares strength of prior beliefs in two states of the world n Posterior odds compares revised beliefs n Symmetrical treatment of competing ideas n Not generally practical to carry out in meaningful situations p Classical n “Null” hypothesis given prominence n Propose to “reject” toward default favor of “alternative” n Asymmetric treatment of null and alternative n Huge gain in practical applicability ™    11/50

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

View Full Document
Part 8: Hypothesis Testing Neyman – Pearson Methodology p Formulate null and alternative hypotheses p Define “Rejection” region = sample evidence that will lead to rejection of the null hypothesis. p Gather evidence p Assess whether evidence falls in rejection region or not. ™    12/50
Part 8: Hypothesis Testing Inference in the Linear Model Formulating hypotheses: linear restrictions as a general framework Hypothesis Testing J linear restrictions Analytical framework: y = X + Hypothesis: R - q = 0 , Substantive restrictions: What is a "testable hypothesis?"  Substantive restriction on parameters  Reduces dimension of parameter space  Imposition of restriction degrades estimation criterion ™    13/50

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

View Full Document
Part 8: Hypothesis Testing Testable Implications of a Theory p Investors care about interest rates and expected inflation: I = b1 + b2r + b3dp + e p Investors care about real interest rates I = c1 + c2(r-dp) + c3dp + e No testable restrictions implied. c1 =b1, c2=b2-b3, c3=b3. p Investors care only about real interest rates I = f1 + f2(r-dp) + f3dp + e. f3 = 0 ™    14/50
Part 8: Hypothesis Testing The General Linear Hypothesis: H0: R - q = 0 A unifying departure point: Regardless of the hypothesis, least squares is unbiased. E[ b ] = The hypothesis makes a claim about the population R – q = 0. Then, if the hypothesis is true, E[Rb – q] = 0. The sample statistic, Rb – q will not equal zero. Two possibilities: Rb – q is small enough to attribute to sampling variability Rb – q is too large (by some measure) to be plausibly attributed to sampling variability Large Rb – q is the rejection region. ™    15/50

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

View Full Document
Part 8: Hypothesis Testing Approaches to Defining the Rejection Region (1) Imposing the restrictions leads to a loss of fit. R2 must go down. Does it go down “a lot?” (I.e., significantly?). Ru2 = unrestricted model, Rr2 = restricted model fit. F = { (Ru2 – Rr2)/J } / [(1 – Ru2)/(n-K)] = F [J,n-K]. (2) Is Rb - q close to 0 ? Basing the test on the discrepancy vector: m = Rb - q. Using the Wald criterion : m (Var[ m ])-1 m has a chi-squared distribution with J degrees of freedom But, Var[ m ] = R [2( X’X )-1] R . If we use our estimate of 2, we get an F[J,n-K], instead. (Note, this is based on using ee /( n - K ) to estimate 2.) These are the same for the linear model ™    16/50
Part 8: Hypothesis Testing Testing Fundamentals - I p SIZE of a test = Probability it will incorrectly reject a “true” null hypothesis.

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

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern