Trend is an important component of a time series. The main methods for determining trend are as follows :

1. Graphical Method: This is an easiest method for determining trend. Taking t on X-axis and time variable yt on Y-axis points are serially joined by linesegments a curve of time series is obtained. A curve passing through close to most of the points of time series curve is showing trend of the given time series. This method is not reliable for determining trend.

2. Method of Least Squares: This method for determining the trend is perfect mathematical and reliable. In this method trend is obtained for each value of t. To fit the equation of linear trend y_t = a + bt, the values of ‘a’ and ‘b’ are obtained such that the sum of the squares of errors [Σe_i² = X(y_i — ŷ_i)² = Z (y – a – bt)²] is minimum.

3. Method of Moving Average: This method determines trend values by eliminating the effect of short-term variations. The period of moving average is obtained by taking the average of oscillations of short term variations. By this method trend is not obtained for each t.