Factor models observe the sensitivity of an asset return as a function of one or more factors. This paper analyzes returns on fourteen stocks of the Croatian capital market in the period from January 2004 to October 2009 using inflation, industrial production, interest rates, market index and oil prices as factors. Both the direction and strength of the relation between the change in factors and returns are investigated. The analyses included fourteen stocks and their sensitivities to factors were estimated. The results show that the market index has the largest statistical significance for all stocks and a positive relation to returns. Interest rates, oil prices and industrial production also marked a positive relation to returns, while inflation had a negative influence. Furthermore, cross-sectional regression with the estimated sensitivities used as independent variables and returns in each month as dependent variables is performed. This analysis resulted in time series of risk premiums for each factor. The most important factor affecting stock prices proved to be the market index, which had a positive risk premium. Keywords: Factor models, risk premium, stock returns, estimated sensitivities, regression analysis JEL Classification: C22, E22, G12 IntroductionFactor or index models observe the sensitivity of an asset return as a function of one or more factors. Depending on the number of factors, there are single-factor or multi-factor models. The basic division of factor models, according to type of factors used, is on the economic and fundamental factor models. Economic factor modDo macroeconomic factors matter for stock returns? Evidence from estimating a multifactor model on the Croatian market els use macroeconomic and financial markets variables as factors, while fundamental factor models use firmspecific microeconomic variables, such as financial indicators. The investment risk can be observed through two components: systematic or market risk, which arises from changes in the macroeconomic environment, and unsystematic risk, that is, the unique risk of an individual asset, which may be reduced or even eliminated through diversification. Factor models focus on systematic investment risk, i.e., the one that cannot be avoided by investment diversification.Factor models are based on the Arbitrage Pricing Theory (APT), introduced by Ross (1976), for more details about APT see for example Campbell, Lo and MacKinlay (1997). Unlike the Capital Asset Pricing Model (CAPM), which estimates the systematic investment risk of an asset by a single factor, the market portfolio, multi-factor models introduce several factors that influence stock returns. However, one of the main limitations of APT arises from the fact that factors, as well as the number of factors, are not known in advance, but they must be determined by statistical or economic analysis.In accordance with the basic division of factor models of stock returns on economic and fundamental, the literature on factor models can also be regarded in this ...

This article studies house price developments in six European countries: Croatia, Estonia, Poland, Ireland, Spain and the United Kingdom. The main goal is to explore the factors driving the rise of house prices in transition countries. Because house price increases in the last two decades are not peculiar to transition countries, the analysis is extended to three EU-15 countries that have recorded house price rises. The similarities and differences between the two groups of countries in terms of house price determinants can thus be explored. In the first part of the empirical analysis VAR is employed to detect how GDP, housing loans, interest rates and construction contribute to real house price variance. In the second part of the analysis multiple regression models are estimated. The results of both methods suggest that the driving forces behind house price inflation in both groups of countries are very similar and encompass the combined influence of house price persistence, income and interest rates.

We introduce a variant of the Barndorff-Nielsen and Shephard stochastic volatility model where the non-Gaussian Ornstein-Uhlenbeck process describes some measure of trading intensity like trading volume or number of trades instead of unobservable instantaneous variance. We develop an explicit estimator based on martingale estimating functions in a bivariate model that is not a diffusion, but admits jumps. It is assumed that both the quantities are observed on a discrete grid of fixed width, and the observation horizon tends to infinity. We show that the estimator is consistent and asymptotically normal and give explicit expressions of the asymptotic covariance matrix. Our method is illustrated by a finite sample experiment and a statistical analysis of IBM™ stock from the New York Stock Exchange and Microsoft Corporation™ stock from Nasdaq during a history of five years.Martingale estimating functions, Stochastic volatility models with jumps, Consistency and asymptotic normality, Trading intensity,

Abstract

Abstract. We provide and analyze explicit estimators for a class of discretely observed continuous-time stochastic volatility models with jumps. In particular we consider the class of non-Gaussian OrnsteinUhlenbeck based models, as introduced by Barndorff-Nielsen and Shephard.We develop in detail the martingale estimating function approach for this kind of processes, which are bivariate Markov processes, that are not diffusions, but admit jumps. We assume that the bivariate process is observed on a discrete grid of fixed width, and the observation horizon tends to infinity.We prove rigorously consistency and asymptotic normality based on the single assumption that all moments of the stationary distribution of the variance process are finite, and give explicit expressions for the asymptotic covariance matrix.As an illustration we provide a simulation study for daily increments, but the method applies unchanged for any time-scale, including highfrequency observations, without introducing any discretization error.