问题描述:
英语翻译
Because we used more than a dozen different measures of real estate activity and wanted to use the same specification for each of the regressions,we sometimes traded statistical significance for specification uniformity.We sacrificed little by using the same specification to explain each measure of either residential or commercial real estate.We settled on a regression specification that explained each of the flows of real estate activity as a function of the changes in personal income and population but of the levels of consumer sentiment,mortgage interest rates,the unemployment rate,the percentage returns on commercial and single-family real estate,and bank capital shortfalls and surpluses.The same explanatory variables were used in each of the regressions reported,with two exceptions:The recent percentage return on commercial real estate and the fixed-rate mortgage (FRM) interest rate were used only in the regressions for commercial real estate,and the percentage change in house prices and the adjustable-rate mortgage (ARM) interest rate were used only in the regressions for residential real estate.
The regression specifications used to generate the results presented in tables 1,2,and 3 differ across columns only in their dependent variables—the measures of real estate market activity.The dependent variables were each expressed in real,per capita terms.The real estate series in tables 1 and 2 were converted to real terms by dividing the nominal values by the level of nominal median house prices in each state.The nominal income series in table 3 were converted to real terms by dividing them by the level of the national consumer price index.In tables 1 and 2,the dependent variables in the first columns of numbers are the changes in the commercial and single-family real estate loans held by commercial banks; the dependent variables in the remaining columns of tables 1 and 2 and in table 3 are the levels of annual flows.
Because we used more than a dozen different measures of real estate activity and wanted to use the same specification for each of the regressions,we sometimes traded statistical significance for specification uniformity.We sacrificed little by using the same specification to explain each measure of either residential or commercial real estate.We settled on a regression specification that explained each of the flows of real estate activity as a function of the changes in personal income and population but of the levels of consumer sentiment,mortgage interest rates,the unemployment rate,the percentage returns on commercial and single-family real estate,and bank capital shortfalls and surpluses.The same explanatory variables were used in each of the regressions reported,with two exceptions:The recent percentage return on commercial real estate and the fixed-rate mortgage (FRM) interest rate were used only in the regressions for commercial real estate,and the percentage change in house prices and the adjustable-rate mortgage (ARM) interest rate were used only in the regressions for residential real estate.
The regression specifications used to generate the results presented in tables 1,2,and 3 differ across columns only in their dependent variables—the measures of real estate market activity.The dependent variables were each expressed in real,per capita terms.The real estate series in tables 1 and 2 were converted to real terms by dividing the nominal values by the level of nominal median house prices in each state.The nominal income series in table 3 were converted to real terms by dividing them by the level of the national consumer price index.In tables 1 and 2,the dependent variables in the first columns of numbers are the changes in the commercial and single-family real estate loans held by commercial banks; the dependent variables in the remaining columns of tables 1 and 2 and in table 3 are the levels of annual flows.
问题解答:
我来补答展开全文阅读