vol. 13 núm. 1 (2020)

The main objective of this paper is to analyze the problems associated with the study of the variables that influence the determination of the technical capacity for the insurance business. In this sense, we tests two methodologies for calculating the total expected losses in a real portfolio of car insurance. In the first step, we modeled a conventional univariate statistical model, where we found that the LogNormal distribution was the best fit. As a second methodology, a copula model was calibrated to incorporate the stochastic association between variables frequency and severity. The results showed that both distribution have extreme behaviors, with a negative and low Kendall correlation (-0.24), but that reject the hypothesis of independence at 5 % confidence. The rotated Clayton 270 degrees copula, with the marginal distributions Exp-Poisson and Log-Normal for frequency and severity respectively, presented the best estimates of the expected loss with a difference of 12 % in relation with the empirical expected loss of the real data base. Finally, we detected that the hypothesis of independence between severity and frequency for this practical exercise leads to significant overestimates of the expected loss.

Recently evidence about the existence of nonlinear structures in time series has been presented in the literature. Often tests are used to contrast linearity vs. nonlinearity. However, the presence of outliers can seriously distort the results (concluding erroneously that is endogenous nonlinearity in the data generating process). Through a simulation study, it is analyze the performance of three tests of nonlinearity when there are outliers. In addition, it is analyze if the series of prices of electricity in Colombia presents a nonlinear endogenous dynamic.

Do you think you spend too much time on your cell phone? For this generation, this device has become indispensable in everyday life. The internet is the main reason for globalization, and these devices, all together contain the best features of innovations and information technologies in a single object, making the cell phone the most popular device used by people around the world. This has produced enormous and evident changes in societies and people, for this reason, a study was done in which the use of cell phones was analyzed, focusing on the students of the Santo Tomas University. This study was carried out using sampling techniques to collect the information and subsequent estimates and multivariate analysis techniques to generate the validation of the survey and the indicators. The main results that were found were indications of addiction, which was manifested in anguish when not using their cell phone for long periods of time, as well as academic benefits under the correct use of the device.

With the aim of studying the statistical procedures used in the postgraduate cour-se in Tropical Agriculture (PAT) of the Universidad de Oriente, period 1998-2016,the statistical management applied in 53 master work presented was analyzed.Variables related to the experimental design, treatment arrangement, number ofrepetitions, characteristics of the experimental unit, types of variables and scalesused, type of statistical analysis, verification of assumptions required by the testsand interpretation of interactions, among others, were identified. The information was systematized in a database and analyzed through descriptive statistics, Binomial test and correlation analysis, according to the variable analyzed. It was determined that the use of statistical techniques differs between mentions of the PAT. The tendency is towards the development of critical experiments with the use of robust experimental designs and treatment arrangements. More than 60 percent of the experiments employ less than four repetitions and there are deficiencies in the description of the experimental units. The assumptions of the ANAVA are not verified and there were deficiencies in the interpretation of significant interactions. The most frequent comparison tests of media were Duncan and Minimum Difference Significant (42.42 and 30.30 percent, respectively), which induce type I error.  

Dear readers, from the editorial team of the journal Communications in Statistics we want to send you a cordial greeting and thank all our authors, reviewers and readers for their support. True to its editorial line, the journal Communications in Statistics continues to present research papers with applications in different areas of knowledge