Solving task on-line

The method of analytic smoothing

Application in the analysis of time series methods of enlarging the intervals and the moving average reveals a trend to describe it, but to get a generalized statistical evaluation of the trend through these methods is impossible. The solution to this higher-order problem - the measurement of trend - is achieved by the method of analytic smoothing.
The main content of the analytical method of alignment in the time series is that the main trend of y _i is calculated as a function of time y_t = f(t_i)

Service will build a model of linear trend of time series y _t , check for residual autocorrelation time series test of Durbin-Watson.

Instructions

Specify the number of data (number of rows), cleck Next. The resulting solution is saved in the file MS Word. For editing, you can use formula editors Microsoft Equation.

Number rows Dependent variable (Confidence interval: if the percentage, the value calculated from the average X) Kind trend Linear y = ax + b By parabola y = ax^2 + bx + c Exponential y = b exp(a) By degree y = b x^a By hyperbola y = a/x + b The number of forecast period 1 2 3 4 5 6 7 8 9 10 Significance level 0.005 0.01 0.025 0.05 0.1 0.25 0.4 Rounding 2 4

Determination of the theoretical (estimated) levels is based on the so-called adequate mathematical function that best displays the basic trend of a number of dynamics.
Selection of the appropriate function is the method of least squares - The minimum sum of squares of deviations between the theoretical y _ti and empirical y _i levels.

Add to favorites

---