This paper applied both univariate and multivariate approaches to forecast the volatility of Swedish Real Estate Sector Index (OMX Stockholm Real Estate PI) from January 1st 2000 to January 1st 2018. OMX Stockholm Real Estate PI is an industry stock index, which contains all shares those are classified as a real estate company and listed on the Stockholm Stock Exchange. This index is constructed based on the guidance of Industry Classification Benchmark (ICB), which is designed to track the performance of real estate industries on NASDAQ OMX Stockholm. 

Firstly, various univariate methods have been applied such as ARIMA, alternative GARCH models to perform the analysis and forecasting. 

Secondly, we analyzed the dynamic correlations of three time-varying indexes: OMX Stockholm Real Estate PI, OMX Stockholm 30 (OMX 30), and Nasdaq OMX Valueguard-KTH Housing Index (HOX Index). OMX30 is one of the Nordic most well-known stock indexes which includes the 30 most traded shares on the Stockholm Stock's Exchange and is considered as the indicator of the performance of the whole Stockholm Stock Exchange. Hox Index is the mostly used housing index in Sweden. Multivariate methods such as GMM, VAR lag model, ARCH models have been applied to investigate the relationship among indexes. 

Finally, we selected OMX Stockholm Real Estate PI, OMX Copenhagen Real Estate PI, OMX Helsinki Real Estate PI, Nordic Real Estate PI and US REITs to study the volatility spillover effects between US and Scandinavian countries and the existence of spillover effects among Nordic countries from period 2000 to 2018. 

As for now, we are working on the model constructions and selections, and the result parts are still in process.