universitybas.blogg.se - Excel student t distribution

Using data from 23 companies on: X - the price of product A, thousand Y - profit of commercial enterprise, study of their dependence is made. Calculation of the correlation index in ExcelĮxample 3. As a result, the hypothesis H 1 about the statistical significance of the coefficient of determination is accepted. Knowing that α = 0.05, p = 2, and n = 53, we obtain the following value for F crit (see Figure 2).

p and n are the numerator and denominator of the degrees of freedom, respectively.

α is the probability associated with a given distribution.We test the hypotheses using the Fisher F-criterion. To test the significance of the equation as a whole, we hypothesize H 0 about the statistical insignificance of the coefficient of determination and its opposite hypothesis H 1 about the statistical significance of the coefficient of determination: To verify the statistical significance of the multiple regression equation using Fisher's F-test, to draw conclusions. Checking the statistical significance of the regression by the FINV functionĮxample 2. Thus, with a probability of 0.95, the linear correlation coefficient lies in the interval from (–0.386) to (–0.990) with a standard error of 0.205. The results of solving this problem with the functions used in the Excel package are shown in Figure 1.įigure 1 - Example of calculations. The standard error of the linear correlation coefficient is calculated.

The interval estimate for the linear correlation coefficient is determined based on the inverse Fisher z-transform.

The interval estimate for a statistically significant linear correlation coefficient is determined.

If the calculated value of t p > t cr, then the hypothesis is rejected, which indicates the significance of the linear correlation coefficient, and hence the statistical significance of the relationship between X and Y. If the hypothesis is confirmed, the t-statistic has a Student's distribution. When testing this hypothesis, t-statistics is used. In this case, the hypothesis about the equality of the correlation coefficient to zero is put forward and verified.

The significance of the linear correlation coefficient is verified on the basis of Student's t-criterion.

Calculate the linear correlation coefficient r xy.

The scheme for solving such problems is as follows: Using data on the activity of commercial organizations, it is required to make an estimate of the relationship between the profit Y (million) and the cost X (million) used for product development (given in Table 1). Evaluation of the relationship of profit and cost function FISHERĮxample 1.