Abstract
We have many kinds of data of time series such as stock prices. We understand time series with a natural language. We have proposed a method to express a global trend and local features of time series in linguistic expressions. For frequent oscillation data, we have proposed a method to express outline of oscillation with expressing global trends of the local maximum and minimum data of difference between the original data and its Triangular Moving Average (TMA). For large oscillation mixed with small oscillation, we cannot have its expected expression.We propose a new method where we choose two local maximum (minimum) data whose degree of slope is the greatest (least) among local maximum (minimum) data in a certain time interval. We form the upper (lower) outline data using these data, from which we calculate the center line of oscillation. And we have its expected expression with difference between the original data and this center line data.
