In global stock markets, it rarely is possible to see a long-run upward - or downward - trend in price movement for an asset, people explain this fact with different interpretations. Using the Improbability Principle as instrument, we investigate if relative to the average this kind of trend can be considered as a random fluctuation. In order to assess the pattern in the market securities we consider a Random Walk, but with a GARCH component to model the variance, with a generalised hyperbolic distribution as conditional distribution to capture the volatility clustering and the heterogeneity of the residual densities. Thus, we show that in a large portfolio it is possible to see an upward movement due to the improbability principle. We estimate the price of the end of the period starting from the year before, with a Monte Carlo simulation based on the estimated residuals of the whole five years. So, we consider the ratio between the number of times the simulated price is greater than the real one and the number of simulations as the probability of having an over-appreciated stock price due to a false movement. In our analysis, only one stock has resulted spurious: the ZAGG company.
On the Improbability Principle in Long-Run Upward Stock Trends
RIBECCO NUNZIATA
;FEDERICO MATTEO
2018-01-01
Abstract
In global stock markets, it rarely is possible to see a long-run upward - or downward - trend in price movement for an asset, people explain this fact with different interpretations. Using the Improbability Principle as instrument, we investigate if relative to the average this kind of trend can be considered as a random fluctuation. In order to assess the pattern in the market securities we consider a Random Walk, but with a GARCH component to model the variance, with a generalised hyperbolic distribution as conditional distribution to capture the volatility clustering and the heterogeneity of the residual densities. Thus, we show that in a large portfolio it is possible to see an upward movement due to the improbability principle. We estimate the price of the end of the period starting from the year before, with a Monte Carlo simulation based on the estimated residuals of the whole five years. So, we consider the ratio between the number of times the simulated price is greater than the real one and the number of simulations as the probability of having an over-appreciated stock price due to a false movement. In our analysis, only one stock has resulted spurious: the ZAGG company.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
