In this paper, an algorithm for estimating all Lyapunov exponents, or Lyapunov spectra of deterministic dynamical systems, is applied to random time series, that is, the time intervals of gamma ray emissions of cobalt. The algorithm used in this paper is that proposed by Sano and Sawada, which estimates Jacobian matrices of an assumed dynamical system from the data points on a possible attractor reconstructed. As a result, the largest Lyapunov exponent of the cobalt data is estimated to be positive, the same as the case of deterministic chaos. Although the sum of all Lyapunov exponents is positive in lower reconstructed dimensions, the values are estimated to be negative when the reconstructed dimension becomes higher. This result is also the same as deterministic chaos in dissipative dynamical systems. It is an indication that naive application of estimating algorithm for Lyapunov spectra to real data does not necessarily lead us to correct results. In order to check the appearance of positive Lyapunov exponents, firstly an analysis by the local versus global (LVG) plots is adopted. The results by LVG plots on the cobalt data are clearly different from those on typical nonlinear dynamical systems in lower dimensional reconstructed state space. Secondly, a direct approach for estimating largest Lyapunov exponents is applied. The results show that estimated Lyapunov exponents on cobalt data do not necessarily indicate existence of positive values. Lastly, a test of statistical hypothesis with surrogate data is also applied to the cobalt data under the null hypotheses that the cobalt data is produced from a linear stochastic system. The results show that statistical significances between the original and surrogate data are so small that some null hypotheses on the cobalt data cannot be rejected; the results imply that an analysis with surrogate data sets could avoid spurious interpretation that random time series, such as the cobalt data, has low-dimensional nonlinear deterministic dynamics with positive Lyapunov exponents.