Forecasting household consumption - A non-linear approach

Detta är en D-uppsats från Handelshögskolan i Stockholm/Institutionen för nationalekonomi

Sammanfattning: This paper aims at improving the forecasting capabilities of the household consumption model currently in use by the National Institute of Economic Research. The mortgage discount, calculated as the difference between the banks' official mortgage rates and the rates that the households actually pay, is added in both a linear and a non-linear fashion as an explanatory variable to an existing model based on the permanent income hypothesis. The result from the estimations are rather mixed and clear conclusions are hard to find. No improvements to forecasting performance are made, but the non-linear model suggests there might be a relationship between a decreasing mortgage rate discount and drops in household consumption whereas an increase in the mortgage rate discount does not seem to lead to an increase in household consumption.

  HÄR KAN DU HÄMTA UPPSATSEN I FULLTEXT. (följ länken till nästa sida)