Affects of Immigration

The impacts of immigration on wages and employment in the US labour market

The impacts of immigration on wages and employment in the US labour market

CHAPTER: METHDOLOGY

Migration in US has increased by a very large amount since late 80s and it is probably considered as the most serious economic issue. The increased immigration is certainly the impact of the unusual inflow of foreign residents on the labour market outcome of natives in US. This has been focused a lot by the economists in U.S. since the last few years. Typically, only minor negative effects of increased immigration have been found in these studies. While there is a strong feeling in the US public that migrants threaten the position of natives in current times of slack labour markets there have been only few econometric studies on this issue with somewhat mixed results.

This particular study incorporates the economic model through various tests because the data contains problems of autocorrelation, multicollinearity, heteroscedasticity and particularly the trends through time series. The regression equation is considered as the last part of the analysis where the analyst assures that there is not problem with the data now.

Initially the data was tested for multicollinearity i.e. if the independent variables are highly depending upon each other or not because this problem leads to insignificant results. For this purpose, the R - square values have been analyzed along with the t - values, if they are significant or not. It happens in multicollinearity that the R - square is high but the coefficient t - values are insignificant. This sign shows that there exists a multicollinearity problem which makes our estimators inefficient and biased and if it is found in the model, remedial measures will be taken.

The model is then tested for autocorrelation and heteroscedasticity, as LM test would be applied for autocorrelation and Lesgjer test would be for heteroscedasticity. Autocorrelation is a problem where the error terms in the model are correlated in a time series data because of similar trends found during years. The problem arises when there is a serial correlation in the variables and it causes the inefficiency in the results. LM test will then be used to detect heteroscedasticity in which the assumption of equal variances in the OLS is violated. Difference in variances disturbs the population results accuracy and this problem is usually found with the improving data collection techniques. Similarly, the test of equality and chow test will be used to see whether the model is stable or not and if the null hypothesis is rejected, we can conclude that the model is stable and does not contains any hikes or usual patterns that can cause the results to be unreliable and biased.

The trends followed in the model in the presence of time series data are detected by cointegration which shows that the trends are significant for the analyses or not. The null hypothesis in the cointegration test is taken as whether the particular trends are significant for the model ...