Abstract

PARODY CAMARGO, Edder; CHARRIS FONTANILLA, Arturo  and  GARCIA LUNA, Rafael. LOG-NORMAL MODEL FOR PREDICTING THE PRICE OF SHARES OF THE BANKING SECTOR. Dimens.empres. [online]. 2016, vol.14, n.1, pp.137-149. ISSN 1692-8563.  https://doi.org/10.15665/rde.v14i1.412.

The following article develops a prediction exercise of share's prices In the banking sector that quoted In the general Index of the Stock exchange of Colombia (IGBC) during the period from 17 to 24 July 2015, using a model Log-normal complemented with Monte-Carlo's simulations, in order to determine goodness of fit test of the model, using the root-mean-square deviation (RMSD). The results indicate that the model is useful to make an approximation to the possible minimal and maximum values that shares can take. Nevertheless, it's results lack the sufficient precision to induce the accurate purchase of this type of financial assets. Since the profitability of these shares is calculated using the last 100 information and the model contributes equal relevancy to data t-100 to t-1. Without mattering if in t-1 moment, the volatility is lower or higher than in the t-100 moment, reason by is recommended in following researches, the application of models with mobile averages of gentle exponential and models of the Arch and Garch family, with major capacity of prediction.

Keywords : log-normal model; actions; volatility; Monte-Carlo simulation; the root mean square error.

        · abstract in Spanish | Portuguese     · text in Spanish     · Spanish ( pdf )