ECON 2504 Study Guide - Final Guide: Autocorrelation, Dorian M. Goldfeld, Time Series

• School: University of Adelaide
• Department: Economics
• Course: ECON 2504
• Professor: Umair Khalil

Document Summary

Objective of slr: fit a straight line in such a way that the sum of squares of the residuals are minimised. [e(x)2]= (cid:523) (cid:523)p*x(cid:524)(cid:524)2= (4/3 + 8/3 + 9/3)2=49. Properties of covariance: if x & y are indep, cov(x,y)=0 (but not vice versa, cov(x,x) = var(x, cov(a+bx,c+dy) = bdcov(x,y) If x & y are independent, then cov(x,y) = 0; Use when population standard deviation is unknown df=n-1, ybar=sample mean, mu=what we are testing t-distribution table: one-tailed is top line two-tailed is bottom (cid:1872)=(cid:3052) (cid:3046)/ (cid:523)don(cid:495)t divide (cid:2009) by 2 when two tailed) If z~n(0,1), z2 is a (cid:2870) with 1 df. (cid:2869)(cid:2870)=(cid:4666) (cid:2869)(cid:4667) (cid:3020)(cid:3118) (cid:3118) is what we are testing for, s2 is the sample. Ho: (cid:2870)= , (a: (cid:2870) or > or < , Ratio of 2 (cid:2870) distributions, each divided by (cid:1858)=s(cid:2869)(cid:2870)(cid:2869)(cid:2870) Bigger standard deviation is always on top (d-1) F distribution df = (n-1) variance their d. f df=n-1: inference and hypothesis testing.