ECON20003 Lecture Notes - Lecture 9: Null Hypothesis, Eviews, Covariance

• School: University of Melbourne
• Department: Economics
• Course: ECON20003
• Professor: Griffiths

Document Summary

Scatter diagrams: graphically depict how two variables are related, negatively/positively and weak/strong, concepts of covariance and correlation measure the strength of a linear relationship. Determines the direction of a linear relationship between two variables. Sample covariance: difficult to distinguish if the covariance value is big/small, magnitude of covariance depends on the units of measurement of the two variables. Correlation: gives a unit-free measure of linear association. Found by dividing covariance by standard deviations of each variable: (cid:883) , (cid:883, 0 correlation means no linear relationship. Two types of correlation coefficients: pea(cid:396)so(cid:374) (cid:272)o(cid:396)(cid:396)elatio(cid:374) (cid:272)oeffi(cid:272)ie(cid:374)t (cid:395)ua(cid:374)titati(cid:448)e data, spea(cid:396)(cid:373)a(cid:374) (cid:396)a(cid:374)k (cid:272)o(cid:396)(cid:396)elatio(cid:374) (cid:272)oeffi(cid:272)ie(cid:374)t o(cid:396)di(cid:374)al data. Testing: null hypothesis is (cid:2868):=(cid:882) (cid:374)o li(cid:374)ea(cid:396) (cid:396)elatio(cid:374)ship between the two variables, alternative can take one of three forms, (cid:2869): (cid:882) (cid:1867) >(cid:882) (cid:1867)