Summary: On the probabilistic space \(( \Omega,\mathrm{F}, \mathrm{P})\), a two-component stationary (in the narrow sense) sequence \(\{\xi_i X_i\}_{i\ge 1}\) is given, where \(\{\xi_i\}_{i\ge 1}(\xi_i : \Omega \to \Xi)\) is a controlling sequence, and the members of the sequence \(\{X_i\}_{i\ge 1}X_i:\Omega\to R\) are observations on some random variable \(X\). The cases of conditional independence and chainwise dependence of these observations are considered. Using observations \(\{X_i\}_{i\ge 1}\), kernel observations of Rosenblatt-Parzen type of an unknown density of the variable \(X\) are constructed. The upper bounds of the mathematical expectations are established for the integral of the standard deviation of the obtained estimates from \(f(x)\).