The Effects of Immigration on the U . S . Economy
tion of previous out-of-sample forecast errors . Given historical data , a sample of such forecast errors is generated by successively applying a chosen point forecasting model to a sequence of fixed windows of past observations and recording the associated deviations of model predictions from actual observations out-of-sample . The suitable quantiles of the distribution of these forecast errors are then used together with the point forecast made by the selected model to construct an empirical prediction interval ( Lee & Scholtes , 2014 , p . 217 ). Vector autoregressive models of GDP growth rate and unemployment rate were used to see overall prediction for 2017-2031 . Additionally , we employ detrending for the dependent variables , which removes a trend from the time series . Considering the removal of an aspect from the data , it may lead to some distortion .
First , in the GDP growth rate prediction , we consider FDI , lawful permanent residents , corporate profits , poverty rate , median family income , urban population growth rate , education , and top 1 % income share . In Figure 8 , the GDP growth rate is linear , increasing above zero from 2017 to 2031 . Also , the results in Figure 4 could be compared to identify the trend of GDP growth rate in stable predictions above zero . Hence , we posit that FDI and immigrants could have a positive impact on the GDP growth rate overall .
Second , for the unemployment rate prediction , we consider FDI , lawful permanent residents , corporate profits , poverty rate , median family income , urban population growth rate , education , top 1 % income share , FDI interaction with the president ’ s party , poverty rate interaction with the president ’ s party , and lawful permanent residents ’ interaction with recession year . Figure 9 shows that the prediction is linear and slowly increasing over 15 years with prediction values between 4 % and 6 %. Comparing Figure 5 in ARIMA , the unemployment rate would be stable between 4 % and 6 %. Moreover , in the empirical forecast intervals model , other variables would be considered to have an influence on unemployment . For example , when a recession occurs , it may impact job opportunities and cause the unemployment rate to increase . On the other hand , the political party in power also impacts foreign investment through policy choices . Earlier , the positive impact that FDI has on unemployment was discussed ; when FDI decreases it may cause the unemployment rate to increase . Overall , we assume unemployment would slowly increase between 4 % and 6 %.
Policy Implications
Figure 10a shows that FDI exists
in relationship with GDP growth rate and urbanization . However , FDI has a negative impact on the GDP growth rate . When there is a change in FDI of $ 1 million dollars , it is associated with a 2.37e-06 decrease in the GDP growth rate . This indicates that FDI has a slightly negative impact . We can see in Figure 10b that in a recession year , FDI is less compared to a non-recession year . Also , we can assume that econom-
69