What is the regression analysis? It is function search that the dependence of some variable on some factors could describe. The equation which will turn out as a result of this research is used to build the line of regression.
It is required to you
- - calculator.
Instruction
1. At first calculate values of signs: factorial and productive (respectively x and y). For this purpose use formulas of the average and simple arithmetic.
2. The equation of regression reflects dependence of the studied indicator on the independent factors influencing it. This equation needs to be found. Its look for a temporary row will represent the trend characteristic of a certain random variable, naturally, in time.
3. In calculations y=ax+b equation is usually used. It is called the equation of simple pair regression. Though is more rare, but nevertheless also other equations are applied: indicative, exponential and sedate functions. As for function type in each single case, he is defined by means of selection of the line that most precisely describes dependence which is investigated.
4. To construct linear regression, it is necessary to determine its parameters. Count them, resorting to the help of analytical programs for the personal computer or the special calculator. It is the simplest to find function elements, having applied classical approach which is based on a method of the smallest squares. The sign has actual values and settlement. And so, this method consists in minimizing of the sum of squares of deviations of the first from the second, and it represents the solution of a system of the normal equations. In a situation with linear regression of a formula on which there are parameters of the equation, are as follows: a = hsr – bxcp; b = ((y*x) sr – usr*khsr) / (x^2) sr – (hsr) ^2.
5. Make on the basis of the data obtained by you regression function now. For this purpose at first calculate average values of variables x and y and set up them in the received equation. So you will find coordinates of points (xi and yi) actually lines of regression.
6. On axis x in the rectangular system of coordinates postpone xi values, for axes y – yi respectively. Too note coordinates of average values. In case schedules are correctly constructed, they will be crossed in such point which coordinates will equal to average values.