Analysis of the Regional Purchasing Power in the Swedish Housing Market : A Multiple Regression Modeling Approach

• Main Authors: Lin, Therese, Maras, Stefan
• Format: Bachelor Thesis
• Language: English
• Published: KTH, Matematisk statistik 2018
• Subjects: Computational Mathematics; Beräkningsmatematik; Valda; Småland; Northern Sweden
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-229735

This report analyzes which macroeconomic factors have the greatest effect on a consumer’s purchasing power on the Swedish housing market. Multiple linear regression analysis is used to construct separate models for each of the eight Swedish regions: West Sweden, South Sweden, Småland and the Islands, Stockholm, East Middle Sweden, Middle Northern Sweden, North Middle Sweden and Upper Northern Sweden. The chosen factors, which are also used as regressors in the regression model, are disposable income, increase in households, employment rate, the Riksbank’s repo rate, square meter price for tenant owned flats and new construction of residential homes. Quarterly data is collected and processed for every macroeconomic factor from the second quarter of 2005 to the fourth quarter of 2015. The affordability is defined as the consumer closing price for tenant owned flats and it is used as the response variable for the regression models. The models are first fitted using all the regressors. To improve the models and potentially eliminate some of the macroeconomic factors a number of different model adjustments, such as Cook’s distance, VIF evaluation, stepwise forward and backward elimination are used. The results show that the purchasing power among the regions is mainly described by disposable income, employment rate and increase in households. Disposable income is the macroeconomic factor that has the greatest effect on the regional purchasing power. Den här rapporten analyserar vilka makroekonomiska faktorer som påverkar hushållens köpkraft på den svenska bostadsrättsmarknaden. Multipel linjär regressionsanalys används för att konstruera separata modeller för Sveriges åtta regioner: Västsverige, Sydsverige, Småland och öarna, Stockholm, Östra Mellansverige, Norra Mellansverige, Mellersta Norrland och Övre Norrland. De valda faktorerna, som också används som förklarande variabler i regressionsmodellen, är disponibel inkomst, hushållsökning, sysselsättningsgrad, riksbankens reporänta, kvadratmeterpriset för bostäder, ...