Section: 3 Market Structure
Sub Section: 3 Time Series Forecast
The Time Series analysis has two main goals: Identifying the nature of a sequence of observations and predicting future values using historical observations (forecasting). In time series analysis it is assumed that the data consist of a systematic pattern and random noise which usually makes the pattern difficult to identify. Most time series analysis techniques use filtering to remove noise.
There are two general components of Time series patterns: Trend and Seasonality. The trend is a linear or non-linear component and it does not repeat within the time range. The Seasonality repeats itself in systematic intervals over time. These two components are often both present in real data.
Trend analysis is a technique used to identify a trend component in the time series data. In many cases data can be approximated by a linear function, but logarithmic, exponential and polynomial functions can also be used. The Dundas Chart supports polynomial approximation and linear approximation (implemented as a special case of polynomial approximation).
Regression Analysis, used in trend analysis, is the study of relationships among variables, and its purpose is to predict or estimate the value of one variable from known values of other variables related to it. Any method of fitting equations to data may be called regression, and these equations are useful for making predictions and judging the strength of relationships.
Forecasting and extrapolation from present values to the future values is not a function of regression analysis. To predict the future time series analysis is used.
To predict values it is necessary to find a predictive function which will minimize the sum of distances between each of the points and the predictive function itself. The least-squares method is the most common predictive function, and it calculates the minimum average squared deviations between the points and the estimated function.