Time Series Forecasting of the Development of the Insurance Industry in Poland

Abstract

This work investigates the development of the insurance industry in Poland over the last twelve years (2001 - 2012) and makes forecasts of this development for all quarters of the year 2013. We consider Gross Written Premiums (GWP) as the best indicator showing the size of the insurance industry. Our aim is to discover relations between GWP and other time series regarding Polish economy: profitability ratio of technical activity for the entire insurance industry, Gross Domestic Product, inflation and consumer confidence indicators. Firstly, we conduct univariate analysis of all the six time series, find trends and seasonal effects, model the residuals, as well as apply SARIMA models. For each series the corresponding forecasts are presented. In the second part we conduct multivariate time series analysis, in particular we model our data with VAR and look for Granger causalities.