27.08.2019 Views

book

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Poster Communications<br />

SP2<br />

Thursday, September 5th<br />

19:45<br />

Functional coefficient regression models for the<br />

problem of Okun’s Law<br />

Michel Helcias Montoril<br />

Sidney M. Caetano<br />

UFJF<br />

In 1962, Arthur Okun posited an empirical relationship between the change in the unemployment<br />

rate and growth rate. He explained the difference in the unemployment rate (Y ) by the growth<br />

rate of the real gross national product (X). Such a relationship became known as Okun’s law.<br />

Using quarter-to-quarter data from 1947 to 1960, Okun fitted a linear regression and concluded<br />

that Ŷ = 0.3 − 0.3X. Over time, Okun’s law has been studied under different models and different<br />

time samples. Among the approaches, it has been considered dynamic models assuming that the<br />

regression coefficient varies over time. In this work, instead of evaluating the regression coefficient<br />

as a function of time, we consider a wavelet-based functional regression model to study the Okun’s<br />

law, where (in our case) the regression coefficient changes according to lagged growth rates of the<br />

GNP. In this case, one can interpret that variations of the GNP in the past will be able to explain<br />

changes in Okun’s law in the present.<br />

Keywords: Functional coefficient regression models; Okun’s law; Wavelets<br />

87

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!