Can Stock Price Fundamentals Properly Be Captured? Using Markov Switching in Heteroskedasticity Models to Test Identification Schemes

Abstract

Structural identification schemes are of essential importance to vector autoregressive (VAR) analysis. This paper tests a commonly used structural parameter identification scheme to assess whether it can properly capture fundamental and non-fundamental shocks to stock prices. In particular, five related structural models, which are widely used in the literature on assessing stock price determinants are considered. They are either specified in vector error correction (VEC) or in VAR form. Restrictions on the long-run effects matrix are used to identify the structural parameters. These identifying restrictions are tested by means of a Markov switching in heteroskedasticity model. It is found that for two of the five models considered, the long-run identification scheme appropriately classifies shocks as being either fundamental or non-fundamental. A series of robustness tests are performed, which largely confirm the initial findings.

Structural Vector Autoregressions: Checking Identifying Long‐run Restrictions via Heteroskedasticity

“…For illustrative purposes two of the models from Velinov () are considered here. Both are trivariate models that have been used widely in the literature.…”

“…In both models changes in volatility are modelled via Markov processes with three states and the same VAR orders as in Velinov (), that is, Model I is a MS(3)‐VAR(2) and Model II is a MS(3)‐VEC model with two lagged differences of in . Modelling changes in volatility by MS models makes sense here because these models are quite flexible by allowing a number of different states and also mixtures of states.…”

Structural Vector Autoregressions: Checking Identifying Long-Run Restrictions via Heteroskedasticity

“…We use the same data as Velinov (2013). In other words, data on GDP, interest rates, stock prices and CPI are from the Federal Reserve Economic Database (FRED) whereas earnings data are from Robert Schiller's webpage.…”

“…In both models changes in volatility are modelled via Markov processes with three states and the same VAR orders as in Velinov (2013), that is, Model I is a MS(3)-VAR(2) and Model II is a MS(3)-VEC model with two lagged differences of ∆y t in (2). Modelling changes in volatility by MS models makes sense here because it is quite flexible by allowing a number of different states and also mixtures of states.…”

Structural Vector Autoregressions with Heteroskedasticity: A Comparison of Different Volatility Models

“…Examples are Lanne et al (2010), Herwartz andLütkepohl (2014), Lütkepohl andNetšunajev (2014a), Netšunajev (2013), Lütkepohl and Velinov (2015), Velinov (2013).…”