The article presents the results of the analysis of the fluctuations of the USD/RUB and the definitions of integrability of the corresponding time series. The research of currency pair by linear methods revealed the presence of non-stationarity in the behavior of the time series, as a result we came to the conclusion that it is impossible to create a high-quality predictive model based on parametric methods.An attempt was made to improve the quality of the forecast to study the integrability of the time series. For this purpose we used various modifications of test the Dickey–Fuller and software products. We tested estimate the order of integrability of two economic time series ,characterizing fluctuations of the U.S. dollar against the ruble and the dynamics of the stock of the Sberbank. The results of the analysis prove that the Dickey-Fuller test is a simple and reliable way to assess the order of integrability of the time series. The benefits of the ADF-test before DF-test are identified.The result is a stationary model of time series , which we can used for forecasting.

Keywords: time series, the main hypothesis, alternative hypothesis, Student criterion, stationarity, nonstationarity, integrability,the test for unit root, the Dickey-Fuller test

`

Keywords: